News Summary:

• NotebookLM is now called Gemini Notebook.

• It remains a separate product rather than becoming only a Gemini app feature.

• Selected paid users can let it write and execute code in a secure cloud environment.

• Google plans to make notebooks available inside AI Mode in Search.

Google has changed the name of one of its best-known AI research tools.

But the change is more than a branding update.

NotebookLM is becoming Gemini Notebook while gaining tools that move it closer to a full research and analysis workspace. Google says the product can now use a secure cloud computer to write and run code based on the sources in the notebook.

Is NotebookLM being shut down?

No.

Google says Gemini Notebook remains a standalone product focused on research. Existing notebooks will continue to work, while users will also be able to access synced notebooks through the Gemini app.

The new name signals that the product is becoming more closely connected to Google’s wider Gemini system.

Google also plans to bring notebooks into AI Mode in Search, although the company has not announced a firm public launch date for that integration.

What Can Gemini Notebook Do With Code?

Each supported notebook can connect to a secure cloud computer. This lets the system create and run code for tasks like cleaning data, comparing figures, finding patterns, and producing charts.

For example, a marketer could upload campaign reports from several countries. Gemini Notebook could help standardize the files, calculate results, and prepare a chart or written report.

Google says the system includes more than 100 curated software skills and can create downloadable files such as the following:

• Excel spreadsheets

• PowerPoint presentations

• PDF and Word documents

• Charts and images

• CSV and JSON data

That makes Gemini Notebook more useful for business research, not just summarizing PDFs or producing audio discussions.

Who Gets the New Features?

The code-execution upgrade is available first to Google AI Ultra users and certain Workspace business customers with eligible AI access.

Google says it will reach Pro users on the web over the following weeks. Availability may therefore depend on the user’s plan, account type, and rollout status.

Consequently, the rename may appear before every account receives the more advanced tools.

Can the Gemini Notebook replace a research team?

Not yet.

Google says the upgraded system performed better than its previous version in internal tests. It reported a 78.2% win rate against its earlier baseline for advanced web research and source discovery.

But that is a first-party comparison. It does not prove that Gemini Notebook is more accurate than ChatGPT, Claude, Perplexity, or a skilled human analyst.

And running code correctly is not the same as choosing the right analysis.

A system can calculate the wrong metric perfectly. It can also produce a polished spreadsheet based on incomplete data or a weak assumption.

Technical note: Grounding the source reduces some errors but does not guarantee that generated code, calculations, or conclusions are correct.

Users should still check formulas, source data, and important business conclusions before publishing or acting on them.

Why This Technical Warning Matters for Marketers and Founders

For agencies, Gemini Notebook could combine several steps that often require separate tools:

1. Gather sources

2. Add campaign or sales data

3. Run an analysis

4. Create charts

5. Produce a report or presentation

Creators and coaches may use it to turn research into teaching materials, worksheets, or client reports.

SaaS founders can use it for competitor research, client feedback analysis, and internal planning.

But affiliate publishers should be careful. Gemini Notebook can help organize public information, yet it cannot replace actual product testing. A generated comparison is not evidence that the writer used the software.

What should you test before relying on it?

Use one real but non-sensitive research project.

Upload trusted sources and a small dataset. Ask Gemini Notebook to calculate known figures, explain its method, and create a downloadable report.

Then verify:

• Every important citation

• Every formula

• Every numerical result

• Every claim not directly found in a source

• Every recommendation based on assumptions

The test will show whether the tool saves real time or simply creates more work to verify.

What Happens Next?

Google says Gemini Notebook: Well over 30 million people and more than 600,000 organizations use Gemini Notebook, according to Google. More than 30 million people and 600,000 organizations use Google Gemini Notebook, according to the company. These figures, provided by Google, have not undergone independent auditing and were included in the announcement. in the announcement.

The larger strategic change is integration.

NotebookLM began as a focused research product. Gemini Notebook is becoming a research layer that can move across the standalone app, Gemini, and eventually Google Search.

That could make it more convenient. It could also make users more dependent on Google’s AI ecosystem.

Gemini Notebook is still NotebookLM at its core. Your notebooks remain, and the product continues to focus on source-based research.

But the addition of code execution, business-file creation, and wider Google integration makes it a more serious analysis tool.

The practical next step is simple: test it against a task where you already know the correct answer. That will tell you more than Google’s benchmark numbers.

News Summary:

• Brands can boost an X post that mentions them.

• Advertisers can add a call-to-action button and destination link.

• Each boost can have its own budget and campaign duration.

• Access requires an X Premium Business subscription.

X is turning organic praise into advertising inventory.

The platform’s new Mentions Boost feature lets a company pay to expand the reach of a post written by another X user when that post mentions the brand. X says marketers can add a custom call-to-action button and link, turning a customer comment into a performance campaign without producing separate creative.

How Does X Mentions Boost Work?

When an eligible brand finds a post mentioning it, the company can choose the boost option, set a budget, and decide how long the promotion should run.

Mentions Boost expands X’s existing Boost product, which was designed to promote an account’s own posts. Social Media Today reports that access requires Premium Business, whose basic plan costs $200 per month before campaign spending.

Why Would Brands Promote Customer Posts?

A genuine customer post may feel more credible than polished advertising copy. It can also reduce the time needed to brief a designer and produce creative and secure internal approval.

But authenticity can weaken once the post becomes paid media.

A positive comment is not automatically an effective advertisement. The campaign still needs suitable targeting, a relevant landing page, and an offer that gives people a reason to act.

The Main Risk Isn’t Creative—It’s Control

X says advertisers are responsible for promoted content and must follow applicable laws, platform rules, and advertising standards. Promoted material is also subject to X’s ad-review process.

The US Federal Trade Commission says endorsements used in advertising must be truthful and not misleading. Its guidance also indicates that highlighting an unsolicited customer review in marketing can turn that review into an endorsement.

Expert flag: The public materials reviewed don’t clearly explain whether the original author must approve the promotion, receive a notification, block it, or what happens when the post is edited or deleted.

Brands shouldn’t treat a visible boost button as permission.

Before promoting a post, please obtain written approval from its author, save the complete original context, verify any product claims, and ensure the destination page matches what the customer actually said.

Is Mentions Boost Worth the Cost?

The feature may suit brands that already receive frequent positive mentions and actively advertise on X.

It’s harder to justify for smaller businesses that must first pay for premium services and then fund the campaign. A company with only a few usable customer posts may get better value from standard ads or creator partnerships.

Run a controlled test. Compare one mention of a Boost campaign with a normal promoted brand post using similar audiences, budgets, and conversion goals.

Measure qualified visits, leads, and sales, not impressions alone.

What Marketers Should Watch Next

The biggest unanswered questions concern author permission, post changes, campaign reporting, and the treatment of negative replies beneath promoted customer posts.

Until X publishes fuller documentation, marketers should treat Mentions Boost as an experimental advertising format not a risk-free shortcut.

News Summary:

• TikTok says it contributed at least £10 billion to the UK economy in 2025.

• The commissioned report estimates that TikTok-linked activity supported 153,000 jobs.

• UK small and medium-sized businesses gained an estimated £3.4 billion in added economic output.

• The figures combine company data, surveys, and economic modeling, rather than direct sales alone.

TikTok says activity linked to its platform added at least £10 billion to the UK economy in 2025, supported 153,000 jobs, and helped small businesses generate billions in extra output.

But the headline is an economic estimate. It is not a count of sales made through TikTok Shop.

What does TikTok’s £10 billion figure include?

The report was led by Public First and includes separate work by EY on TikTok’s direct spending in Britain.

Public First estimates that users, creators, and businesses using the platform added more than £9 billion to the economy. When TikTok includes its own operations and investment, the combined estimate rises to at least £10 billion.

The report describes the figure as gross value added, or GVA. This measures the value created by economic activity after subtracting the goods and services used to produce it.

It is not the same as TikTok’s revenue, merchant sales, or money paid directly to creators.

Public First says the total was equal to about 0.3% of UK gross domestic product. It also states that TikTok-linked activity supported 153,000 jobs.

A separate survey result says businesses using TikTok hired about 31,000 people for work connected to the platform. Those figures should not be added together because they measure different things.

Small businesses are central to TikTok’s argument.

TikTok says UK small and medium-sized businesses gained £3.4 billion in added economic output during 2025.

In Public First’s business survey, 84% of companies using TikTok said the platform had helped increase sales or revenue. Twenty-eight percent said they hired extra staff whose work included TikTok.

TikTok Shop is also a big part of the company’s UK growth story.

TikTok says more than 300,000 small businesses now sell through the service. It reports more than 6,000 LIVE shopping sessions each day and says that LIVE shopping grew 55% from the previous year.

These figures show why short video is moving beyond brand awareness. A TikToker can describe a product and let a viewer purchase it by TikToking a separate online store.

But sales are not profitable.

A seller may still pay creator commissions, platform fees, discounts, delivery costs, and refunds. High gross sales can lose money once those costs are counted.

TikTok says online discovery also drives offline spending.

The report estimates that 6.1 million people visited a local café or restaurant after discovering it on TikTok. It says 4.3 million visited an independent shop.

It also links the platform to 2.3 million additional visits to UK towns, cities, and attractions.

In publishing, TikTok-related discovery was estimated to generate £17,000 in extra book spending. Music activity, including royalties, tickets, and merchandise, was valued at £430 million.

These findings point to a wider change in local marketing.

A short video about a café, shop, or tourist site can work like a search result, customer review, and personal recommendation at the same time.

For small firms, such content may lower the cost of reaching new customers. A business does not always need a large existing audience if one useful or entertaining video reaches the right TikTok.

How is the £10 billion estimate calculated?

The method matters because TikTok commissioned the research.

Public First says it began with TikTok’s filed turnover and estimated the UK share of advertising revenue using market data. It then used a return-on-ad-spend measure from NielsenIQ to estimate business revenue linked to TikTok advertising.

The researchers converted that revenue into GVA and used official UK economic tables to estimate indirect effects.

Those effects include activity supported elsewhere in supply chains and the wider economy.

The study also used surveys of UK adults and businesses. Survey answers can show what people believe influenced a purchase or visit, but they cannot prove that every action would not have happened without TikTok.

So the report is evidence of economic influence—not a bank statement for Britain.

What should businesses measure instead?

The £10 billion figure helps TikTok show its scale to advertisers, regulators, and policymakers.

It does not tell an individual company whether its TikTok activity is profitable.

Businesses should track incremental revenue—sales that would not have happened without the campaign. They should also count production costs, creator payments, advertising spend, discounts, fulfillment, returns, and repeat purchases.

Local firms should measure store visits, bookings, and branded searches before and after TikTok activity.

E-commerce sellers should calculate the contribution margin for each product after platform, creator, and delivery costs.

Public First estimates that casual creators on platforms such as TikTok could help drive another £27 billion in UK e-commerce growth by 2030. That is a modeled forecast, not a guaranteed result.

The report’s clearest message is narrower: social discovery can lead to sales, jobs, and real-world visits.

But businesses still need their numbers.

A national estimate may justify paying attention. Only company-level profit can justify continued spending.

Yes, getting people to actually see and buy what you sell.

Rachel Rofe’s program focuses on selling physical products through live shows on the Whatnot app, rather than requiring you to build an online store, learn paid advertising, create a large social following, or wait months for search traffic.

That sounds appealing, especially for beginners.

But a $2,497 training decision deserves more than excitement and income screenshots. I wanted to understand what the system includes, how the business model works, what it realistically demands from a new seller, and where the main risks sit.

So, I bought the founding member offer and now sharing my complete unbiased review of QuickAudience.com

Quick Audience Profit System Review at a Glance

• Name: “Quick Audience” WhatNot Profit System

• Training format: Guided paid program

• Product sourcing: Dedicated tools and strategies

• Show planning: Scripts, structure and software

• Coaching: Weekly calls and help desk

• Community: Private member group

• Beginner support: High

• Additional costs: Course, inventory and equipment

• Best for: Buyers wanting guidance

• Official SIte: QuickAudience.com/join

• Short Quick Audience Review: The Quick Audience mentorship program is best suited to beginners in Whatnot and supports countries that want structured training, sourcing tools, AI apps, coaching, and a clear launch success path and process. Its primary advantage is convenience and guided implementation, but its value depends on whether you will consistently run live shows, manage inventory, fulfill orders, and use the support instead of only watching the training.

  

  Check the official website for discounts:

What Is Rachel Rofe’s Quick Audience Profit System?

Rachel Rofe’s “Quick Audience” Profit System is a mentorship program built around live selling on the WhatNot app. Rachel has 10+ years of e-commerce, dropshipping, and print-on-demand business model experience. And, right now in 2026, she comes up with a new idea that works better than others!

The Quick Audience business model is relatively simple:

• Find products people want.

• Create a scheduled Whatnot livestream.

• Present or auction the products in real time.

• Package and ship completed orders.

• Make happy customers come back to buy more.

• Test new products, restock, and repeat.

The best thing about this approach is its simplicity. You do not need to build a traditional Shopify store, master Amazon FBA, wait for Etsy approvals, create complex funnels, learn paid ads, or spend months building an audience before testing whether people want your products.

However, “simple” does not mean passive.

A serious seller still needs to research products, secure inventory, prepare shows, speak or interact live, track sold items, package orders, manage customer expectations, and keep returning even when the first few shows are quiet.

How Rachel Rofe’s Whatnot Strategy Works

The system turns Rachel Rofe’s broader live-selling process into a repeatable workflow.

Her approach centers on six stages:

1. Source suitable products.

2. Plan the order and structure of the show.

3. Run the live show.

4. Ship purchases efficiently.

5. Nurture repeat buyers.

6. Restock based on what sells.

This is more practical than simply telling beginners to “go live and start selling.”

The real value is in reducing uncertainty. A beginner usually becomes stuck on questions such as the following:

• What should I sell?

• How much should I pay for inventory?

• What price should I start an auction at?

• How many products do I need?

• What should I say during the show?

• How do I avoid losing money after fees?

• What happens if nobody joins?

• How do I ship dozens of orders correctly?

The Quick Audience WhatNot Mentorship Program is designed to answer those questions through training, tools, coaching, and implementation support.

What You Get Inside Quick Audience Profit System

Now let’s look at what’s inside the program. Actually, when I purchased the founding member plan, I got 8 core products and 7 special bonuses. Let me describe each in detail.

the app

1 - The Master Blueprint

The core training program teaches beginners how to create an account on the Whatnot app and set up a seller to run live shows.

I believe the course should be the starting point for someone who has never sold on Whatnot. Instead of trying to assemble a complete process by watching random YouTube videos, members receive the steps in a specific order.

The practical benefit here is speed and reduced confusion. You are paying $2.5K to avoid months of scattered research and unnecessary mistakes.

2 - Profit Radar

I really like this “Profit Radar” app, which is a product-evaluation tool. It’ll help us to determine whether an item is worth buying, how quickly it may sell, and what price could leave room for profit.

The app is one of the most commercially interesting parts of the offer. Nad, you may find this very appealing!

As a WhatNot seller, we need to do product sourcing. This process is where many resale businesses succeed or fail. A bad purchase can tie up capital, create unsold inventory, and force the seller to reduce prices later.

A tool like this improves buying decisions and potentially saves more money than the course costs. But buyers should still apply judgment rather than treating every software recommendation as guaranteed demand.

3 - Opportunity Hunter

In my opinion, “Opportunity Hunter” helps us to identify product categories based on our budget, experience, interests, and comfort level while selling in this app.

This app comes with features that are useful for beginners who understand the Whatnot opportunity but have zero idea what niche to enter and make profits in.

Choosing a profitable niche is not just about finding high-priced products. A strong niche should ideally have:

• Repeat purchase potential

• Reliable sourcing

• Sufficient profit margin

• Active buyer interests

• Products that are easy to present live

• Manageable shipping requirements

Replenishable niche-based items can be more valuable than one-time unique ones. Because repeat customers prefer related niche-based items, and as you know, they are the major part of a sustainable live-selling business.

4 - Show Architect

The “Show Architect” is intended to help sellers like us who have zero experience in this method to plan the flow of a live show.

This plan matters more than it may initially appear.

A live show can become disorganized quickly when the seller is trying to present products, answer comments, run auctions, track buyers, introduce the next item, and maintain momentum at the same time.

A structured show plan can reduce awkward pauses and help the seller decide:

• Which products to open with.

• When to introduce higher-value items.

• How to maintain energy.

• How long should each auction run?

• When to use fixed pricing.

• How to transition between products.

• How to close the show.

For a nervous beginner, this type of structure can make the difference between launching and continuing to delay.

5 - Profit X-Ray Pro

We can calculate potential profit after costs such as platform fees by using the Profit X-Ray Pro tool.

I believe it’s one of the essential apps we need to have. Because revenue screenshots do not tell you whether a seller is actually making money in the Whatnot app.

A proper calculation needs to consider:

• Product acquisition cost.

• Marketplace commission.

• Payment-processing charges.

• Packaging.

• Seller-funded shipping adjustments.

• Giveaways.

• Promotions.

• Refunds or cancellations.

• Unsold inventory.

• Labor and preparation time.

A WhatNot seller can generate impressive gross sales and still earn a disappointing hourly return. This Profit X-Ray Pro tool solves the major problem that many other courses and opportunity-style programs underemphasize.

6 - Weekly Seller Lab Calls

As quick audience members, we get access to weekly coaching calls where we can ask questions and receive feedback from experts.

What do you think about it? Is it more valuable than the recorded lessons?

In my opinion, this type of support is what more buyers desire in the current AI era. Real human support and guidance can make a difference in the outcome.

Rachel’s training videos explain general principles in a step-by-step manner. And the live coaching call helps us to understand a specific product, niche, show, pricing concern, or sourcing decision. And mentors guide us with a unique blueprint on what works now in the WhatNot app.

The value will completely depend on whether you attend consistently and ask focused questions. Paying for a coaching program and having the systems but never using them is one of the easiest ways to make a program like this feel overpriced.

7 - Rapid Help Desk

We receive private support with responses targeted within 48 hours. In my first support query, I got a reply within 4 hours. Maybe I’m lucky, or maybe it’s because there are fewer numbers of support tickets.

Such help desk support is really beneficial for us to reduce implementation delays. Beginners may otherwise spend days trying to solve a small issue that prevents them from launching. Furthermore, we’re getting expert-level guidance with the support desk team.

Support matters most when it is

• Specific rather than generic questions.

• Current with Whatnot policies.

• This is relevant to the seller’s country and category.

• It is focused on action and roadblocks.

• When a real problem occurs.

8 - Inner Circle Community

We also get a private community of like-minded members in one place to share questions, discuss ideas, ask for feedback, guide others, and win.

This is a useful community that can provide social proof and accountability altogether. Seeing other beginners launching and succeeding may make it easier for us to move past our hesitation.

There is also a strategic benefit with such a community approach. Whatnot sellers can support and direct viewers toward one another, creating potential for collaboration inside the member group, which will be excellent. And I believe together we could grow much faster and better.

In my opinion, there is a limitation: community quality depends on how many active members are participating each day. We need to understand that a private group is not automatically valuable simply because it exists. So, we need to spend some time every day in this community to make it a better place to hang out.

Quick Audience Special Bonus Packages

Other than these 8 core offers, Rachle really shocked us by offering 7 special bonuses that are really useful for each member.

Bonus 1 - 30 Day Behind-the-Scenes Access

As a Quick Audience member, we get the opportunity to observe how Rachel Rofe’s Whatnot strategy works in a real demo, how she does all the activities, and how she sources, prices, runs shows, and responds to problems.

This bone bonus is more valuable than any other. Because live selling contains demo session details that are difficult to communicate through a theory class alone.

Watching an experienced seller doing this business model may be helpful for us to understand pacing, language, product presentation, pricing, and buyer interaction.

Bonus 2 - Fast Start Launch Pad

This is one of my favorite bonuses in this series. Actually, the Fast Start Launch Pad guides us to the shortest route to get our first show live rather than consuming every lesson before taking action.

As you may be aware, most coaching program buyers spend weeks learning but never launch. This unique bonus solves a common course problem. As a member, we could get a return for our investment back quickly this way.

Bonus 3 - Brick-and-Mortar Gold Mine Plan

Rachel’s raining guide to partnering with local retailers that already have inventory. She has a unique plan for making this partnership.

By following this plan, as a seller, we can potentially help those businesses move products through Whatnot and receive a fee or percentage commission rather than buying all the inventory personally.

I believe this approach is a more advanced strategy, but it may appeal to people with limited startup capital or strong local business relationships.

Bonus 4 - Free Inventory Unlocker

Here we get scripts and approaches for securing products through consignment or revenue-sharing arrangements.

The inventory cost is one of the largest barriers to entering a physical-product business. But with these arrangements, the free inventory unlocker bonus is really useful for many of us.

However, “free inventory” does not mean no work or no responsibility. You still need agreements, trustworthy partners, clear tracking, pricing rules, and a process for handling unsold goods.

Bonus 5 - Go Live! Safety Net

When it comes to this scenario, the eGo Live Safety Net bonus is intended to reduce our mistakes and uncertainty before a live stream show begins.

This feature is useful for most buyers of this program who feel overwhelmed by technology or fear making an embarrassing mistake during their first live session.

Bonus 6 - Scroll Stopper Suite

I really like this Scroll Stopper Suite bonus. It really helps with show titles, descriptions, images, and promotional copy. And it can be useful for other social media platforms’ content creations as well. I’m planning to start using the same approach with TikTok and Instagram to get the same reach for my short videos and reels.

Even in the WhatNot app, a live show still needs to earn the attention of viewers. But in WhatNot, each viewer is a potential buyer, and even with just 10 people, it can make a big difference in our bank account. Using better titles and visual presentation can improve the likelihood that browsing users stop and enter. This bonus is really useful and worth it for me!

Bonus 7 - Hot Category Alerts

This bonus is designed to notify members about emerging product trends. With hot category alerts, we are aware of what the current trending items will be. Additionally, it is beneficial to take advantage of cashing in on that trend.

Yes, such trending information can be commercially useful for many of us, but timing really matters here. A hot category can become saturated quickly, and sellers should never purchase large amounts of inventory only because an alert describes a product as promising.

You need to utilize these alerts and check the competition as well.

So, with these factors, 7 bonuses make the Quick Audience program worth the try. And I highly recommend you order this mentorship program and try it out. Because there is a 60-day money-back guarantee, it is not working for you. Just email the support and get your investment back.

Selling on Whatnot: The Reality Behind the Opportunity

One of the strongest reasons to consider Whatnot is that it combines shopping and entertainment in a central hub.

Buyers are not only comparing static listings or product catalogs. They are watching products in real time, asking questions, joining auctions, interacting with sellers, and sometimes developing better relationships with other viewers.

That creates several advantages:

• Faster feedback

• Real-time buyer interaction

• Urgency from timed auctions

• Potential bidding competition

• Repeat customers

• Organic platform discovery

• No separate e-commerce website required

But the same format creates additional demands as well.

Your First Live Shows May Be Quiet

In our first class, Rachel Rofe mentioned shows that generated little or no revenue before momentum was possible. There may be 0 or 1 or 2 live views in your initial live show.

This pattern happened to other active WhatNot sellers as well. Some built meaningful sales only after running many shows, including sessions that attracted few viewers or produced no purchases.

That is an important reality check.

The Quick Audience Profit System may improve the learning curve, but it cannot eliminate the need for repetition, patience, and consistency. Only when we consistently follow this process does the accelerated momentum occur.

Live Selling Is Not Passive Income

We should address this point before moving on to other details. I would like to share this information with you. It’s crucial to understand that live selling is not a passive income-generating machine!

As a Whatnot seller, we may need to

• Source inventory

• Inspect products

• Organize and number items

• Create listings

• Prepare the filming area

• Host the live show

• Answer comments

• Track every sale

• Package purchases

• Print labels

• Ship orders

• Resolve mistakes

• Restock

A 60- to 90-minute live show can require additional hours of preparation and fulfillment. But it’s worth it!.

Quick Audience System buyers should need to treat this method as an active resale business, not an automated income stream.

Repeat Buyers Matter

Most of the successful WhatNot sellers repeatedly emphasize customer service, quick shipping, communication, and trust.

The strongest Whatnot businesses do not depend entirely on finding a new customer for every transaction. They encourage buyers to return to future shows.

That is why a replenishable category can be powerful. Collectibles, accessories, children’s clothing, coins, toys, and similar products may generate repeat demand more naturally than a one-time household purchase.

You Do Not Necessarily Need to Show Your Face

One of my common concerns is camera anxiety. I’m naturally an inborn “introvert.”

So, my first consideration before purchasing this program is whether I can make this business model work. In my research I figured out some sellers build strong businesses while showing only their hands and products. Even in Rachle’s case study interviews, I found two people doing the selling by just showing their hands. Furthermore, other sellers use a rotating display, voice-only interaction, or a faceless setup.

Thus, it’s possible for an introvert and camera-shy person to also make it happen.

But some categories may benefit from showing your face to help build a connection and trust. It’s not a requirement, but naturally when we make profit with a faceless identity, we may try our real face to make more money.

Whatever the mode you apply, face-to-face or faceless, you still need to communicate clearly, present products honestly, and keep viewers engaged.

Quick Audience Profit System Pros and Cons

So, now we’ve reached the area where I can share what I like and dislike about this business model and the program.

What I Like

• The business model is easier to test than many traditional e-commerce models

• The offer combines training, software, coaching, support, and community

• Profit calculation receives more attention than in many revenue-focused programs

• The sourcing tools address one of the most difficult parts of reselling

• Beginners receive help in planning their first show instead of being told simply to take action

• The program supports an extended time and not only prerecorded videos

• The no-inventory and consignment approaches can help to reduce the initial financial risk

• Whatnot handles the payment processing and shipping labels

• A seller can begin without an existing social audience

• Faceless selling is possible

Where I Dislike It and It Falls Short

• The purchase price is significant for a beginner

• The course fee is not the complete startup cost

• Whatnot seller eligibility is limited by country

• Results depend heavily on consistent live shows

• Physical-product fulfillment creates ongoing work

• Sourcing tools cannot guarantee that an item will sell profitably

• No training program can manufacture buyer demand for a weak product

• Some shows may produce little or no revenue, particularly at the beginning

• Inventory mistakes and fulfillment errors can reduce profit

• The model is unsuitable for anyone wanting passive or anonymous automated income

How Quick Audience Profit System Compares to Shopify and Amazon FBA.

Now let’s compare this Quick Audience program and system with other traditional methods.

Quick Audience vs. Shopify

Shopify gives you control over your website and brand, but it does not automatically provide customers.

A new Shopify owner typically needs the following:

• Advertising

• Social media

• Search traffic

• Influencer partnerships

• Email marketing

• Conversion optimization

• Additional apps setup

Whatnot takes away the burden of building a website and generating traffic, as there are already active shoppers on the marketplace. The trade-off is being platform-dependent and having to sell live.

Quick Audience vs. Amazon FBA

Amazon FBA can offer massive marketplace demand, but it often requires more inventory capital, product research, listing optimization, advertising, and account-management knowledge.

That can be faster to test, as sellers can start with personal items, thrifted items, consignment items, or small inventory quantities.

Amazon might be better for regular scalable products. Whatnot might be better for sellers. If you like interaction, fast turnover, collectibles, unique goods, and live demos.

To Whom Is the Quick System Best For?

The program is best suited for:

• Beginners seeking a structured WhatNot success roadmap

• Resellers moving from eBay, Poshmark, Etsy, or Facebook Marketplace

• People who are comfortable speaking during live shows

• Camera-shy sellers willing to use a hands-only setup

• Buyers who will attend coaching calls

• People with access to products, thrift stores, collectors, retailers, or consignment opportunities

• Sellers who can package and ship orders quickly

• Buyers looking for an active side business rather than passive income

• People in countries supported by Whatnot seller registration

Who Should Avoid It?

The Quick Audience Profit System is not the right fit for everyone:

• Anyone expecting guaranteed income

• People unwilling to run live shows consistently

• Buyers who cannot access a supported Whatnot seller account

• Anyone unable to manage physical products and shipping

• People who hate talking to customers

• Buyers who can’t afford the program without financial hardship

• Anyone seeking a fully automated online business

• Experienced Whatnot sellers who already have strong sourcing, show planning, and fulfillment systems.

Is the Quick Audience Profit System Actually Worth It?

In my opinion, the Quick Audience mentorship program is worth the investment when three conditions are met:

1. You can legally and practically sell through Whatnot.

2. You are committed to running multiple live shows.

3. You will actively use the software, coaching, support, and community.

The program is less attractive when someone buys it primarily for motivation.

At a premium price, its value must come from faster execution, fewer sourcing mistakes, improved show performance, and access to knowledgeable support.

A buyer who launches quickly and avoids even a few expensive inventory mistakes could receive meaningful value. A buyer who watches the lessons without running shows is unlikely to see a return.

My Power-Packed Quick Audience Bonus

I will update the bonuses soon!

Frequently Asked Questions

Is the Quick Audience Profit System worth the money?

The Quick Audience Profit System can be worth the money for a committed beginner who wants structured training, sourcing assistance, live coaching, and implementation support. It is unlikely to be worth the price for someone who prefers self-directed learning or is uncertain about hosting live shows.

Are there hidden costs?

Yes, the cost is not just the cost. Depending on your approach, you may need inventory, packaging materials, a label printer, lighting, storage, a phone stand, promotional spending, and enough working capital to operate before payouts arrive.

Who should avoid the Quick Audience Profit System?

People outside supported Whatnot seller countries, buyers seeking passive income, anyone unwilling to ship physical products, and people who cannot comfortably afford the program should avoid purchasing it.

Final Verdict

The “Quick Audience” WhatNot profit system and mentorship program is not a magic button, and that is exactly why it may appeal to serious buyers.

It provides a structured approach to a real business model: finding products, presenting them live, generating sales, fulfilling orders, building repeat customers, and improving through consistent action.

The standout advantage is not one individual tool. It is having the training, product research, show planning, profitability calculations, coaching, support, and community all organized in one place.

The central question is not whether live selling on Whatnot works. Sellers are already demonstrating that it can be done.

The real question is whether you are prepared to run the shows, learn from slow sessions, serve buyers properly, and treat the process as a business.

For an eligible and action-oriented beginner, Rachel Rofe’s Quick Audience Profit System offers a more guided route than trying to figure out every step alone.

News Summary:

• On July 12, 2026, YouTube and FIFA held their first-ever Creator Cup in New York City.

• The event was led by creator IShowSpeed and streamed on FIFA’s and IShowSpeed’s YouTube channels.

• Participating creators had more than 270 million combined subscribers.

• The event formed part of a wider FIFA plan to use YouTube and creators to reach younger, digital-first fans.

YouTube and FIFA did more than stage a creator football match in Central Park.

They built a distribution network.

The first YouTube FIFA Creator Cup brought together online creators, athletes, football figures, and brand partners for a live event at Wollman Rink in New York City on July 12.

The match was streamed worldwide through FIFA’s official YouTube channel and simulcast on the channel of creator IShowSpeed. The creators involved had more than 270 million subscribers between them, according to YouTube.

That number does not mean 270 million people watched the event. It represents the creators’ combined subscriber bases.

But it shows why the format matters.

Instead of relying on one television audience or one official social account, FIFA and YouTube connected the event to many separate creator communities at the same time.

The Creator Cup was built for distribution, not just entertainment.

Traditional sports events usually depend on a central broadcaster.

A creator-led event works differently.

Each participant arrives with an audience, a style of communication, and a group of fans who may not follow FIFA’s official accounts. When those creators publish clips, reactions, behind-the-scenes videos, and short-form posts, the event can travel far beyond the original livestream.

YouTube said the Creator Cup featured players from more than 40 countries. IShowSpeed served as a captain, while football figures and digital creators joined the wider event.

Dove Men+Care and Lay’s were named as livestream partners.

This created four connected layers of distribution:

1. FIFA’s official channels

2. YouTube’s platform promotion

3. The creators’ individual audiences

4. Sponsor and media amplification

That structure is more important than the final score.

It turns one live event into many pieces of content that can continue reaching viewers after the match ends.

Why FIFA wants creators close to the World Cup

The Creator Cup was not a stand-alone experiment.

It is part of a wider agreement that made YouTube a preferred platform for the 2026 FIFA World Cup.

Under that agreement, World Cup broadcast partners can publish more highlights, behind-the-scenes footage, short videos, and selected live match content on YouTube. FIFA also promised greater access for a worldwide group of creators.

Some rights-holding broadcasters can stream the first 10 minutes of matches on their YouTube channels. The goal is to attract viewers and direct them toward full coverage through official broadcasters.

The arrangement gives FIFA a way to reach people who may discover football through a creator rather than through a sports network.

The Associated Press has reported that FIFA is using streaming platforms and digital personalities to connect with younger audiences who expect more informal and interactive coverage.

That shift does not mean television has stopped mattering.

It means major sports groups no longer expect one screen, one channel, or one style of coverage to reach everyone.

Creators give sports events a different kind of reach.

Creators do not simply provide large follower counts.

They provide context.

A creator can explain an event in familiar language, show parts of it that official coverage misses, and speak directly to a specific community. It can help make a big world event feel more intimate.

It can help FIFA turn casual watchers into active fans.

For YouTube, it creates more viewing before, during, and after matches.

And for creators, official access can produce content that would be difficult to make without a platform or sports partner.

But raw subscriber totals should be treated with care.

Many fans may follow more than one participating creator. Some accounts may be inactive. And only part of each audience will see or watch the content.

So a combined reach figure is not the same as unique viewers.

The biggest marketing lesson: access is becoming media inventory.

Sports sponsorship once focused heavily on signs, television spots, and logo placement.

Creator access adds another form of value.

A brand can now appear in livestreams, creator posts, short clips, interviews, and behind-the-scenes content. The event becomes a source of media that can be cut into many formats.

That gives sponsors more ways to appear in front of audiences.

But it also makes measurement harder.

A marketer studying a campaign like the Creator Cup should not stop at subscriber counts or video views. The campaign should be measured across several areas:

• Unique viewers

• Livestream watch time

• Peak concurrent viewers

• Repeat views

• Creator-audience overlap

• Branded search growth

• Social mentions

• Clip performance

• Website visits

• Sales or sign-ups

Without that data, a large reach claim can sound more useful than it really is.

What creators and brands can learn from the event

The Creator Cup offers a repeatable model for other industries.

A software company, online course business, or e-commerce brand may not be able to build a global football event. But it can still use the same structure.

Start with one clear event or story.

Bring in creators whose audiences have a real reason to care.

Give each creator material that fits their normal style instead of forcing everyone to publish the same advertisement.

Then plan content for before, during, and after the event.

The event itself may last one hour. Its useful life should be much longer.

A practical campaign model

A smaller business could apply the Creator Cup approach through the following:

• A live product challenge

• An expert roundtable

• A creator competition

• A public build session

• A virtual workshop

• A community-led product launch

The key is to treat creators as distribution partners, not rented advertising space.

That means involving them early, giving them room to tell the story in their own voice and measuring what each partnership actually produces.

Expert flag: Combined follower counts can hide audience overlap.

The most important number in a creator campaign is rarely the total number of followers.

Ten creators may each have one million followers, but many of those people may follow several of them. Adding all the follower counts together can therefore overstate the real audience.

Brands should ask platforms or agencies for estimated unique reach, audience overlap, and watch-time data.

They should also separate awareness results from business results.

A video can reach millions of viewers without producing meaningful website visits, leads, or sales.

Why this matters for YouTube

YouTube is competing for a larger role in live sports without trying to copy traditional television in every way.

Its strength is not only the livestream.

It is the network around the livestream: search, recommendations, creator channels, Shorts, highlights, comments, and on-demand viewing.

The Creator Cup showed how those parts can work together.

A fan might first see a short clip from a creator, watch part of the live event, search for a player, and later view highlights on FIFA’s channel.

That path is harder to build through a single broadcast.

What happens next?

The larger test will be how YouTube and FIFA use creators during the rest of the 2026 World Cup.

The main questions are not limited to total views.

Publishers and marketers should watch:

• Whether creator coverage drives viewers to official matches

• Which content formats hold attention longest

• How much access creators receive

• How sponsors are included

• Whether viewers trust creator coverage

• How FIFA measures sales, subscriptions, and long-term fan growth

The Creator Cup has shown the model.

The World Cup will show whether that model can work at full scale.

The YouTube FIFA Creator Cup was more than a promotional football match.

It was a test of how a sports organization can combine official media, creator audiences, platform distribution, and sponsor support in one campaign.

The strongest lesson is simple: audience size alone is not the strategy.

The strategy is building a system in which many trusted voices can carry the same event into different communities, then measuring whether that attention creates lasting value.

News Summary:

• TikTok will test stronger detection of accounts dedicated to publishing AI-generated spam.

• The first tests will focus on politics, current events, financial advice, and medical content.

• A new in-app learning hub will teach users how to spot AI-generated media.

• TikTok says it has labeled well over 3 billion videos using several AI-detection and disclosure methods.

TikTok is tightening its response to low-quality AI content, but it is not banning creators from using artificial intelligence.

The company said it will test improved systems for finding accounts that repeatedly publish AI-generated spam. It is also expanding its content-labeling program and launching educational tools to help users understand what they are watching.

The changes were detailed in a TikTok Newsroom announcement published July 10.

TikTok’s new test targets accounts built around AI spam.

TikTok said the test will focus on accounts “dedicated” to posting AI-generated spam rather than every account that uses AI.

The company did not publish a full definition of an AI spam account. It also did not disclose how much content an account must publish before its activity is treated as spam.

The first stage will cover subjects that can affect public trust or personal well-being:

• Politics and current events

• Financial advice

• Medical and health content

These areas carry greater risks because false or misleading information may influence elections, financial decisions, or health choices.

TikTok says it removed well over 86 million fake accounts during the first three months of 2026. That number is a company-reported enforcement figure and has not been independently audited.

This is not a ban on AI-generated TikTok videos.

The announcement does not say that TikTok will remove all AI-made content or punish creators merely for using an AI tool.

TikTok continues to develop its own AI creation features, including tools called Smart Split and AI Outline. The company is also testing controls that let users adjust how much AI-generated content appears in their feeds.

The new enforcement effort focuses more narrowly on mass-produced spam that may crowd out original creators or create risks in sensitive subjects.

That distinction matters.

A creator using AI to edit an original video is not necessarily doing the same thing as an anonymous network publishing hundreds of repetitive financial or medical clips. TikTok’s challenge will be separating legitimate creative work from automated content operations designed mainly to capture views.

A new AI literacy hub is coming to TikTok.

TikTok has worked with the National Association for Media Literacy Education and AI specialist Henry Ajder on a guide for using and understanding AI tools.

The platform plans to launch an in-app learning hub in the coming weeks. It will appear when users search for terms related to artificial intelligence and will provide practical guidance on identifying AI-generated content.

TikTok has not announced a firm launch date or confirmed whether the hub will become available in every market at the same time.

The company also supports AI education from organizations including NoFiltr and the Raspberry Pi Foundation. TikTok says content produced through this program has received well over 200 million views since November 2025 and that it has committed more than $4 million to the broader effort.

TikTok is expanding its use of content credentials.

TikTok is also joining the steering committee of the Coalition for Content Provenance and Authenticity, better known as C2PA.

C2PA maintains an open standard called Content Credentials. These credentials can carry a record of how a photo, video, or audio file was created and edited.

TikTok began using the standard in 2024 to recognize some AI-generated material uploaded from other platforms. When supported information is found in a file, TikTok can apply an AI-generated label automatically.

TikTok now says it has labeled more than 3 billion videos as AI-generated content. It uses a combination of the following:

• Content Credentials

• Labels added by creators

• TikTok’s detection systems

• Invisible watermarking technology

The figure is cumulative. It does not show how many labeled videos were misleading, how many labels were applied automatically, or how accurate each detection method was.

Expert warning: An AI label is not a fact check.

A label can tell viewers something about how content was made. It cannot prove that the message inside the video is true or false.

The C2PA technical explainer states that Content Credentials verify whether recorded origin information is properly formed, linked to the file, and free from tampering. They do not judge the truth of the content itself.

That creates two important limits:

First, an AI-generated video may contain accurate information.

Second, a video without an AI label is not automatically human-made, accurate, or trustworthy. Credentials may be missing because the original tool did not add them, a platform did not preserve them, or the file passed through software that does not support the standard.

Content labels should therefore support fact-checking and media literacy—not replace them.

Creators and marketers should take the following actions now.

Creators do not need to stop using AI. But accounts that rely on high-volume automation should review their publishing process.

Label realistic AI content clearly.

Use TikTok’s AI disclosure tools when a video contains realistic AI-made people, voices, places, or events. Do not depend only on captions such as “made with AI,” especially when TikTok provides a formal label.

Add clear human value.

AI-assisted videos should contain original reporting, testing, commentary, experience, or creative work. Changing a few words in the same template and publishing it repeatedly is more likely to resemble spam.

Sensitive topics may require further review.

Finance, medicine, politics, and current events now receive more attention.

Claims, sources, and disclosures should be reviewed by a qualified person before being published by businesses. An AI system should not be the final authority on whether advice is safe or correct.

Audit automated publishing systems

Teams using scheduling tools, AI video generators, or automated content pipelines should verify the following:

• How many videos are being published each day?

• Whether multiple videos repeat the same claims

• Whether sources are visible and credible

• Whether required AI labels are preserved

• Whether a real person reviews the final output

And brands should keep records of the AI tools used to create or edit each campaign asset. That makes corrections easier if a label is missing or a piece of content is challenged.

What this means for creators

TikTok’s announcement may help creators whose original work competes with large amounts of cheap, repetitive content.

But much depends on how accurately TikTok identifies spam accounts. Automated enforcement can produce false positives, especially when legitimate publishers use templates, translation tools, or repeatable educational formats.

TikTok has not explained whether suspected accounts will be removed, have their reach reduced, or receive a warning first. It has also not described an appeals process for this specific test.

Creators should therefore avoid assuming that any change in reach is connected to the new system unless TikTok provides an account notice or further documentation.

What this means for marketers and agencies

For brands, the main issue is trust.

An unlabeled AI spokesperson, false product demonstration, or mass-produced advice video can damage a company even when an outside agency or contractor created it.

Marketing teams should add AI disclosure checks to their approval process. Contracts with creators and production agencies should also state who is responsible for labeling AI-generated material and keeping records of the tools used.

There is no evidence in TikTok’s announcement that correctly labeling AI content will automatically reduce its reach. Hiding AI use may create a larger compliance and reputation risk than disclosing it.

What remains unclear

TikTok has confirmed the direction of the test, but several important details are still missing:

• How the platform defines an account “dedicated” to AI spam

• Which countries will receive the detection test first?

• Whether enforcement will involve removal, reduced distribution, or warnings

• How TikTok will handle false detections

• Whether creators will receive a specific appeal process

• When the AI literacy hub will launch in each region

These details will determine whether the program mainly removes large spam networks or also affects ordinary creators using AI in legitimate workflows.

TikTok is not drawing a simple line between human content and AI content.

It is drawing a line between disclosed, original creative work and scaled output that looks like spam—particularly when that output deals with health, money, politics, or breaking news.

For creators and marketers, the safest response is straightforward: disclose meaningful AI use, publish less repetitive content, verify sensitive claims, and make sure a real person remains accountable for every post.

News Summary:

• YouTube launched Gifts in India on July 7, 2026.

• Viewers buy Jewels and use them to send animated Gifts during eligible live streams.

• Creators receive Rubies, which YouTube converts into earnings.

• Creators must be in the YouTube Partner Program, be at least 18, and have a verified AdSense account.

• Enabling Gifts revokes access to Super Stickers during live streams.

YouTube has brought its virtual gifting system to India, adding another way for eligible creators to earn directly from their live audiences.

The feature allows viewers to buy a digital currency called ‘Jewels’ and spend it on animated gifts. These animations appear over a creator’s live stream, while the creator receives Rubies that represent earnings. YouTube announced the India launch on July 7.

How YouTube Gifts Work

The system uses three separate virtual items:

• Jewels are the digital credits viewers purchase.

• Gifts are the animations viewers choose and send during a live stream.

• Rubies are the units creators receive when viewers send those gifts.

YouTube says viewers can buy jewels in bundles, which means they do not need to complete a new payment each time they want to send a gift. The selected animation then appears as an overlay on the live video.

The company has also created India-themed animations, including Vada Pav, Pani Puri, and Chai Toast. Other options use familiar expressions such as “Badhai Ho,” “Kem Cho,” “Macha,” and “All Izz Well.” YouTube says it plans to add more seasonal gifts later.

How Much Can Creators Earn in Rubies?

YouTube’s policy documents provide a clearer earnings formula than its launch announcement.

A creator earns one Ruby for every two Jewels viewers redeem for Gifts. YouTube currently values each Ruby at $0.01 in net revenue, and the money is paid with the creator’s other YouTube Partner Programme earnings. Creators can view estimated Rubies and related revenue in YouTube Analytics.

That does not mean a viewer’s Jewel purchase price directly affects a creator’s payout. The price viewers pay for Jewel bundles and the amount creators receive are separate parts of the system.

Indian creators should also avoid treating the dollar figure as a guaranteed rupee payout. The final amount received may be affected by currency conversion, taxes, payment rules, and the creator’s AdSense account.

Who Is Eligible for YouTube Gifts in India?

The feature is not automatically available to every YouTube channel.

To earn from gifts, a creator must

• Be part of the YouTube Partner Program.

• Be at least 18 years old.

• Have a verified AdSense account or another form of age verification accepted by Google.

• Live in a country where gifts are available.

• Accept YouTube’s Virtual Items Module.

India now appears on YouTube’s official list of countries where creators can enable gifts and where viewers can buy jewels.

Creators who meet the stated requirements may still need to wait for the option to appear in their account. YouTube says eligible channels will find Gifts inside the Earn section of YouTube Studio.

How Creators Can Turn On Gifts

Eligible creators can turn on the feature in YouTube Studio.

On a computer:

1. Open YouTube Studio.

2. Select Earn from the left menu.

3. Open the Supers & Gifts tab.

4. Find “Turn on Gifts in live streams.”

5. Select Get Started.

6. Review and accept the Virtual Items Module.

On mobile, creators can open the Earn section in YouTube Studio, select Gifts, and tap “Turn on”. YouTube says Gifts will then be enabled automatically on supported live streams.

Not Every Live Stream Can Receive Gifts

YouTube excludes several types of streams from virtual item monetization. You are connected

Gifts cannot be used when a stream is

• Age-restricted.

• Made for children.

• Unlisted or private.

• Running with live chat disabled.

• Connected to a YouTube Giving fundraiser.

Creators must also continue to follow YouTube’s monetization help rules, community guidelines, copyright policies, and AdSense requirements.

Expert Flag: Gifts Are Not Classified as Tips

YouTube explicitly states that virtual items are not crowdfunding, donations, or tipping tools.

That distinction is relevant for creators who describe viewer payments, prepare tax records, or make claims about fundraising. Creators should use YouTube’s own terms—Jewels, Gifts, and Rubies—rather than presenting the feature as a formal donation system.

Enabling Gifts Removes Super Stickers

Creators should understand one important trade-off before activating the feature.

YouTube says channels that enable Gifts will no longer have access to Super Stickers on live streams. The company’s India announcement does not say that Super Chat, memberships, or Super Thanks will be removed.

Creators should therefore compare how much engagement and revenue they currently receive from Super Stickers before changing their setup.

This is especially important for channels with audiences that already understand and regularly use YouTube’s older fan-funding tools.

A Documentation Conflict Creators Should Know About

YouTube’s current help pages are not fully consistent about supported live-stream formats.

Some official pages say gifts work on eligible vertical and horizontal live streams. However, another YouTube eligibility page still lists horizontal live streams among the formats where gifts are unavailable.

The India announcement simply refers to “live streams” and does not resolve the difference.

Until YouTube updates the conflicting pages, creators should check whether the Gifts control appears for their planned stream format and run a private production test before promoting the feature to viewers.

Why This Matters for Indian Creators

YouTube is trying to move livestream monetization beyond advertising.

Ads depend heavily on watch time, demand from advertisers, reactions, and YouTube’s revenue-sharing system. Gifts allow viewers to support individual moments during a live broadcast, such as a performance, announcement, reaction, or question-and-answer session.

That could be useful for gaming creators, educators, entertainers, coaches, and community-led channels that already attract active live audiences.

But gifts should be treated as an extra revenue source, not a predictable replacement for advertising, memberships, brand deals, or affiliate income. Earnings will depend on how many viewers are willing to purchase Jewels and use them during a stream.

YouTube says the number of Indian channels earning seven figures or more in annual rupee revenue increased by over 20% year over year. It also says 92% of Indian creators surveyed agreed that YouTube helps them build a strong audience community.

The first figure comes from YouTube’s internal data, while the second comes from an Ipsos survey commissioned for creator research; neither should be treated as an independent measure of what a typical creator earns.

What Indian Creators Should Do Now

Creators should first open the Earn section of YouTube Studio and check whether the Virtual Items Module is available.

Before enabling it, review recent Super Sticker earnings, confirm that the linked AdSense account is verified, and decide which live broadcasts are most likely to produce meaningful viewer interaction.

Then test the feature during a normal live session rather than building a major event around it immediately.

Track four numbers after each stream:

• The number of viewers who sent gifts.

• Total Rubies earned.

• Revenue per live viewer.

• Whether the gift activity changed watch time or chat participation.

That data will show whether Gifts adds real value to the channel or simply replaces an existing fan-funding method.

What Remains Unclear

YouTube has not publicly explained how quickly Gifts will appear across every eligible Indian account.

The company also has not provided India-specific Jewel bundle prices in its launch announcement, nor has it published typical creator earnings for the country.

And because YouTube’s support pages currently conflict over horizontal live streams, creators should not assume that every stream format will work until they confirm it inside Studio.

YouTube Gifts gives eligible Indian creators a new way to turn live audience participation into revenue.

The setup is straightforward, but the business decision is not. Creators should check eligibility, understand the Ruby payout system, and compare Gifts with their existing Super Sticker performance before enabling it across every livestream.

The feature is most likely to help channels that already have active communities. It will not create that community on its own.

News Summary:

• Cloudflare and OpenAI announced the research pilot on July 8, 2026.

• The test will use signals such as content freshness, real page changes, and traffic quality.

• The companies say the goal is better crawling, indexing, and answer accuracy.

• The signals are not confirmed ranking factors and do not guarantee a ChatGPT citation.

Cloudflare and OpenAI are testing a new way for AI search systems to find useful web pages faster.

The project could matter to publishers, marketers, and online businesses that want their latest work to appear in AI answers. But this is still a research pilot—not a new ranking system or a promise of more referral traffic.

What Cloudflare and OpenAI announced

Cloudflare announced the pilot on July 8. Participating websites may provide network-level signals that show when content is fresh, when a page has truly changed, and whether traffic appears to be of high quality.

OpenAI will use its search systems and real user queries to study whether those signals can make web discovery more efficient. Cloudflare says more than 20% of the web sits behind its network, giving it a broad view of how pages and traffic change over time.

AI search tools need current information, but crawling the web repeatedly takes time and computing power. A signal pointing to a real update could help a system revisit the right page sooner.

“Indexing” means storing web content so a search system can find it later. “Crawling” means visiting pages with automated software to read them.

Why publishers should pay attention

For publishers, the possible value is the chance to show an AI system that a page has changed in a meaningful way.

A news article may change after a court filing. A software review may need a new price. Search tools commonly use signals such as links, sitemaps, and page data to discover new or updated content.

The pilot will examine whether Cloudflare’s network can offer clearer and faster evidence that an important change has occurred. Its stated aim is to improve the accuracy and timeliness of AI-generated answers.

But Cloudflare and OpenAI have not said that a recently updated page will rank higher, appear in ChatGPT, or receive a source link. The announcement also does not provide a public list of participating websites or a timetable for wider access.

That difference matters.

A faster crawl is not the same as a citation. And a citation is not the same as a click.

What website owners should do now

Website owners should not change every publication date or make tiny edits merely to look fresh. That can weaken trust when a page has not gained useful information.

A better approach is to keep technical signals accurate. Use honest publication and update dates. Maintain an XML sitemap. Keep canonical tags correct. Ensure important pages are not blocked by mistake. Add a visible update note when it helps readers understand what changed.

OpenAI says OAI-SearchBot is used to surface websites in ChatGPT searches. A publisher can allow that bot while separately blocking GPTBot, which OpenAI uses for content that may support model training. OpenAI says a robots.txt change may take about 24 hours to affect its systems.

Publishers can also measure visits from ChatGPT. OpenAI says links from its search results include the “utm_source=chatgpt.com” tag, which tools such as Google Analytics can track.

Cloudflare already offers crawler controls.

The pilot sits beside Cloudflare’s existing AI Crawl Control tools.

These tools allow website owners to examine AI crawler activity, block selected crawlers, monitor robots.txt behavior, and use Pay Per Crawl in supported cases. Cloudflare also provides analytics showing how AI crawlers interact with pages on a domain.

This creates a difficult balance for publishers.

They want their work discovered and cited, but many also want control over model training, large-scale scraping, and payment. Allowing an AI search bot does not require giving every crawler the same access.

Publishers should therefore set rules based on purpose. Search discovery, model training, and page visits triggered by individual users are different activities, even when they come from the same AI company. OpenAI’s documentation treats these activities as separate crawler categories.

What remains unclear

The largest unknown is whether the signals will create a measurable gain for publishers.

Cloudflare says the test focuses on answer accuracy and timeliness. It does not promise publishers more traffic, control over citations, or payment for content.

The pilot could help OpenAI use its crawling resources more efficiently. But that alone would not solve the business problem faced by websites when AI summaries answer a question without producing a visit.

There is also no confirmed evidence that the method will become available to every Cloudflare customer or that other AI search companies will adopt similar signals.

What publishers should do next

Publishers should treat the project as an important experiment, not a new SEO rule.

The safest plan is still to publish original reporting, make updates that add real information, keep crawler settings clean, and measure AI referrals separately from traditional search traffic.

Cloudflare and OpenAI may be building a faster road between fresh web pages and AI-generated answers.

Whether that road sends meaningful traffic back to the people who created the content is still the most important question.

News Summary:

• Meta launched Muse Image on July 7.

• One feature let users reference public Instagram accounts with an @mention.

• Meta removed that option on July 10 after receiving criticism.

• Muse Image and its other editing tools remain available.

Meta has removed a new artificial intelligence feature that allowed people to use public Instagram accounts as references when generating images.

The reversal came only days after the company introduced Muse Image, its first image-generation model from Meta Superintelligence Labs.

What did Meta remove?

Muse Image was launched as a tool for creating and editing pictures inside Meta AI. It can follow written prompts, combine photos, remove objects, and place readable text inside images.

At launch, Meta also said users could @mention a public Instagram account. The system could then use public photos from that profile as visual references for a new AI-generated image.

On July 10, Meta updated its announcement and said the public-account reference option was no longer available.

The company said it had intended to provide a useful creative tool and to give account holders control over whether they could reference their content. But Meta acknowledged that the feature had “missed the mark” after receiving feedback.

Why did the feature cause concern?

The main issue was consent.

According to TechCrunch, users were not designed to receive an alert when someone referenced their public photos through the feature. Critics warned that people might not know when someone used their pictures in an AI creation.

Talent agencies and entertainment groups also objected. The Los Angeles Times reported that Creative Artists Agency, United Talent Agency, and performers’ union SAG-AFTRA called for clear opt-in permission before a person’s image or likeness could be used.

An opt-in system keeps a feature disabled until a person chooses to activate it. Meta’s original system instead gave eligible public-account users an option to turn the feature off.

That difference matters because a public photograph is not automatically permission to place someone in a new or misleading setting.

Muse Image has not been withdrawn.

Meta has removed only the ability to reference public Instagram accounts through an @mention. The muse image itself remains available.

The model continues to power image creation in Meta AI, Instagram Story effects, and image generation in some WhatsApp chats. Meta also plans to expand it to Facebook, Messenger, and advertising tools.

So this decision is not a retreat from AI-generated media. It is a narrower reversal involving how public social-media content can be reused.

What marketers and creators should learn

Brands should not treat a public post as automatic approval for commercial AI use.

Before using a person’s face, name, or content in an AI-generated image, marketers should obtain clear permission and keep a record of that approval. They should also verify whether the final image could suggest an endorsement that the person never gave.

Creators should continue watching platform settings and policy updates. A removed feature can return in a different form, and controls may vary by country or account type.

Meta has not said whether it will redesign the account-reference feature or introduce a stricter opt-in version.

The key question is whether future AI tools will ask for consent before using identifiable people—not after complaints begin.

For agencies and creators, the safest rule is simple: public does not mean permission.

The real question for businesses is not which model is “best.” It is the model that is worth paying for.

News Summary:

• OpenAI says GPT-5.6 is now available across ChatGPT, Codex, and the OpenAI API.

• The model family has three main tiers: Sol, Terra, and Luna.

• Sol is the flagship model. Terra is the balanced option. Luna is the fastest and cheapest option.

• Luna API pricing starts at $1 per 1 million input tokens, and Sol API pricing goes up to $5 per 1 million input tokens.

• Rather than always using the most powerful model for every situation, businesses should choose the model that best suits the task at hand.

OpenAI Releases GPT-5.6 With Three Model Tiers

OpenAI has released the GPT-5.6 model family for broader use, with three main versions: Sol, Terra, and Luna. The company describes Sol as its flagship model, Terra as a lower-cost model for everyday work, and Luna as its fastest and most affordable model.

The rollout matters because OpenAI is no longer selling one simple idea of “the best model.” It is pushing users toward model selection based on cost, speed, and task difficulty.

That is a practical shift for marketers, creators, agencies, SaaS founders, affiliate operators, and small teams. If you use AI for research, writing, coding, spreadsheets, landing pages, support replies, product analysis, or automation, the wrong model choice can quietly raise your costs.

What Is GPT-5.6 Sol?

GPT-5.6 Sol is the highest-capability model in the GPT-5.6 family.

OpenAI says Sol is designed for demanding work, including professional analysis, complex coding, computer use, long-running workflows, cybersecurity tasks, design work, presentations, documents, and spreadsheets. The company says Sol reached 90.4% on BrowseComp and 62.6% on OSWorld 2.0, two tests related to browsing and computer-use tasks.

For a small business, Sol is the model to consider when the task is complex and errors are expensive.

That could include:

• Financial analysis

• Legal-style document review

• Complex spreadsheet building

• Technical debugging

• Multi-step market research

• Building a polished slide deck

• Reviewing a messy business workflow

• Creating or testing code

But Sol is also the most expensive GPT-5.6 tier in the API. OpenAI lists Sol at $5 per 1 million input tokens and $30 per 1 million output tokens.

So Sol should not be the default model for every small task.

What Is GPT-5.6 Terra?

GPT-5.6 Terra is the middle option.

OpenAI describes Terra as a lower-cost model with performance competitive with GPT-5.5. It is meant for everyday work where users still need strong quality but may not need the full cost of Sol.

For most content and marketing teams, Terra may become the practical default.

Use Terra for:

• Blog outline generation

• First-draft newsletters

• SEO briefs

• Email sequences

• Product descriptions

• Social post variations

• Basic customer-support drafts

• Research summaries

• Landing-page structure

• Competitor note-taking

Terra costs $2.50 per 1 million input tokens and $15 per 1 million output tokens through the API, according to OpenAI.

That price difference matters. If an agency is producing hundreds of drafts, briefs, summaries, and internal documents every week, Terra may deliver enough quality without paying Sol-level rates for every pass.

What Is GPT-5.6 Luna?

GPT-5.6 Luna is OpenAI’s fastest and most affordable GPT-5.6 model.

OpenAI positions Luna as the low-cost tier for high-volume or lighter tasks. It is priced at $1 per 1 million input tokens and $6 per 1 million output tokens through the API.

Luna is not the model you would choose for a critical legal memo, a difficult coding issue, or a high-stakes financial model.

But it may be the right choice for tasks such as tagging content or cleaning short text.

• Tagging content

• Cleaning short text

• Summarizing simple notes

• Rewriting headlines

• Creating quick social captions

• Classifying support tickets

• Drafting simple replies

• Formatting content

• Generating product-title variations

• Running low-risk automation steps

For affiliates and e-commerce operators, Luna could be useful for repetitive content operations. Think product data cleanup, meta descriptions, category text, search-query grouping, or review-summary extraction.

GPT-5.6 Pricing: The Simple Business View

Here is the practical pricing difference from OpenAI’s API rates:

Model Best Use: GPT-5.6 is ideal for challenging, high-value work. GPT-5.6 Sol costs $5 per 1 million tokens for input and $30 per 1 million tokens for output. GPT-5.6 Terra has a cost of $15 per million tokens and $2.5 per million tokens. LunaFast offers low-cost tasks at a rate of 1 token per 1 million tokens for input and 6 tokens per 1 million tokens for output.

The cost gap is not small.

Sol output costs five times more than Luna output. That does not mean Luna is “better value” for every task. It means teams need a routing strategy.

A useful rule: start with the cheapest model that can reliably complete the task, then upgrade only when the risk or complexity justifies it.

What Should Creators and Marketers Use?

Creators and marketers should not automatically use Sol for every piece of content.

If you just need to plan basic content, write social media captions and email subject lines and repurpose content, Luna or Terra may be enough. If you want more in-depth product comparisons, editorial analysis, paid campaign strategy, or stronger newsletter essays, Terra is probably the best place to start.

Use Sol when the work needs deeper reasoning, sharper judgment, or several steps of analysis.

For example, if you are writing a detailed software comparison, use Terra for the first structure and Sol for the final editorial review. When turning social-search data into an owned content plan, you could start with Terra and then use Sol for the strategic layer.

OnlineCOSMOS has already covered how creators can use Google Search Console to track social content on Google Search. That kind of query data becomes more useful when paired with the right AI model for research, clustering, and content planning.

What Should Agencies and SaaS Teams Use?

Agencies and SaaS teams should think in workflows, not prompts.

A smart workflow may use all three models:

1. Luna for sorting, tagging, formatting, and simple summaries.

2. Terra for drafting briefs, emails, documentation, and support replies.

3. We have successfully sorted the final strategy, complex analysis, coding, and client-facing deliverables.

The software stops teams from wasting money on routine tasks but keeps quality high where it matters.

It also creates a cleaner review process. Junior work can be run through Luna or Terra. Senior-level judgment can be reserved for Sol and human review.

The Big Warning: Benchmarks Are Not Your Workflow

OpenAI reports strong results for GPT-5.6 across coding, professional work, computer use, cybersecurity, academic tasks, and multimodal tests. But those are still vendor-reported results unless repeated by independent testers.

That matters.

A benchmark can show model strength, but it does not prove the model will perform well in your exact business process.

Axios also noted that the independent review of GPT-5.6 was still limited during the early public rollout, with much of the early excitement coming from early testers and OpenAI-linked discussions.

So don’t buy the model story. Test the workflow.

What Is Still Unclear?

A few important questions remain.

We do not yet know how GPT-5.6 will perform across a wide range of small-business tasks once more independent users test it. We also do not know how many companies will actually save money after adding model routing, prompt caching, agent workflows, and human review.

Reuters reported that OpenAI’s broader GPT-5.6 launch followed a limited-access period connected to U.S. government concerns about national security risks from powerful AI models.

That background is important, but it does not change the practical question for most businesses: how do you use the right model without wasting money or increasing risk?

GPT-5.6 is not one model decision. It is a model-routing decision.

For small businesses, the simplest starting point is this:

• Use Luna for cheap, repeatable, low-risk work.

• Use Terra for everyday marketing, content, support, and research.

• Use Sol for complex, high-value work where mistakes cost more than the model.

The best AI setup will not be the one that always uses the strongest model.

It will be the one that uses the right model at the right step, with human review where judgment still matters.

Turn on your camera. Show a product. Start an auction. Watch the bids rise.

But Whatnot is not a normal online store with a video added to it. It is part store, part social network, and part live show.

That mix can help the right business sell products quickly and build loyal customers in the Netherlands. It can also waste time, shrink profit margins, and create a stressful shipping mess.

So, is Whatnot right for your business?

2026?...These eight truths will help you decide before you buy more inventory or plan your first show.

1. Whatnot is entertainment before it is e-commerce.

A normal online shopper searches for a product, compares prices, and checks out.

A Whatnot shopper may open the app without knowing what they want. They stay because the host is interesting, the products are surprising, or the auction feels exciting.

That means you are not only selling an item.

You are also selling the experience of watching you sell it.

The strongest hosts explain products, answer questions, welcome returning viewers, and keep the show moving. They make buyers feel like they are part of a small community.

As one live commerce founder explained,

“What live commerce really cracks: connection.”

That quote appeared in a 2026 Vogue report about live shopping.

This is why a great product does not always create a great Whatnot show. The seller must make the product interesting in real time.

What this means for you: Whatnot is a strong fit when you enjoy teaching, demonstrating, storytelling, or speaking with customers.

2. The growth is real, but the big numbers can be misleading.

Whatnot says sellers produced over $8 billion in live sales in 2025. It also says well over 20 million new accounts were created that year.

According to the company’s 2026 State of Live Selling report, sellers who stream three or four times per week average more than $13,000 in monthly sales.

Those numbers sound exciting.

But remember three things.

First, this is Whatnot’s own report. It is useful, but it is not an independent study.

Second, an average can be pulled upward by a smaller group of very large sellers. It does not tell us what a typical beginner earns.

Third, sales are not profit.

A seller can make $13,000 in sales and still have weak profits after paying for the following:

• Inventory

• Platform fees

• Packaging

• Employees

• Giveaways

• Damaged or lost products

• Time spent preparing and shipping orders

Do not ask only, “How much can I sell?”

Ask, “How much will I keep after every cost?”

3. Some products are much better for live selling.

Whatnot works best when shoppers want to see, discuss, compare, or collect the product.

Strong live-selling products often have at least one of these qualities:

• They are visual.

• They are limited or difficult to find.

• Each item is a bit different.

• Condition is important to buyers.

• The seller can show the items to potential buyers.

• Customers enjoy collecting them.

• There is a story behind the product.

This phenomenon is why categories such as trading cards, comics, coins, vintage clothing, sneakers, jewelry, toys, and collectibles often fit the format.

You can display a vintage jacket from all perspectives. A seller can explain its history, point out flaws, answer sizing questions, and compare it with the next jacket.

That is harder to do with a plain product that looks the same in every store.

For example, a standard phone cable may sell better through search-based e-commerce. Most buyers do not need to watch a live show to understand it.

A simple test: Can you discuss each product for one minute without repeating the product description?

When the answer is yes, the product may work well on Whatnot.

4. High sales do not protect weak profit margins.

Whatnot does not charge sellers to create or store listings. However, it takes fees when products sell.

In the United States, Canada, and Australia, many categories have an 8% commission. Payment processing is commonly 2.9% of the total order value plus $0.30 per transaction. Some categories and regions have different rates.

You can review the current numbers on the official Whatnot seller fees page.

Fees are only one part of the calculation.

Imagine that you buy an item for $12 and sell it for $20.

The $8 difference is not your final profit. You must still account for selling fees, packaging, labor, discounts, giveaways, and any shipping costs you agree to cover.

Auctions add another risk.

When only a few buyers are watching, an item may sell near its opening price. Starting everything at $1 can create energy, but it can also turn valuable inventory into a loss.

A safer beginner strategy is to set the opening price near the lowest amount you are truly willing to accept.

Excitement is not a business model.

5. Your personality becomes part of the product.

You do not need to be loud, funny, or highly confident.

You do need to be present.

Live shoppers notice when a host seems bored, confused, unprepared, or focused only on pushing the price higher.

Good hosts tend to do a few simple things well:

• They greet viewers.

• They describe products clearly.

• They answer questions honestly.

• They keep track of what is selling.

• They admit when they do not know something.

• They continue speaking when the room is quiet.

This process can be tiring.

A normal product listing works while you sleep. A live show needs your attention for most of the time it is running.

You must also prepare before the stream and pack orders after it.

A two-hour show may become four or five hours of total work once you count product setup, listing, cleanup, customer messages, and shipping.

Whatnot may be a poor fit when you strongly dislike performing, answering repeated questions, or following a regular show schedule.

6. The real work begins after the camera turns off.

A quick show can produce dozens or even hundreds of orders.

That sounds like success until you must match every buyer with the correct item, choose the right box, confirm the weight, print labels, and ship everything on time.

Whatnot tells sellers to drop packages off with the carrier within two business days of the sale.

The company provides a full shipping checklist for new sellers.

This operational work is easy to underestimate.

Before running a large show, ask:

• Where will sold items be placed?

• How will each item be matched to its buyer?

• Who will pack the orders?

• Do you have enough boxes and envelopes?

• Please let me know if everything can be shipped within two business days.

• What happens when an item is missing or damaged?

A smaller-audience seller with clean shipping can build a better long-term business than a seller with a large show and constant order mistakes.

7. Fast auctions can encourage bad decisions.

The power of live selling is its speed. A short countdown builds the tension. Others bid, and viewers watch. They may be afraid of losing the product. It’s just a few taps to buy. It can drive sales, but it also creates an ethical imperative. In 2026 Vogue reported that quick auctions, rewards, loud streams, and short buying windows could cause consumers to buy on impulse.

Sellers do not need to remove all excitement. Fun is part of the format.

However, it is important to distinguish between excitement and pressure.

• I need additional information on what the outcome will be.

• Claiming an item is rare when it isn’t

• Misrepresentation of retail value to buyers

• Hiding damage or product defect

• Shaming those viewers who don’t bid

• Exploiting false urgency

• Persuading individuals to go overboard

• Offering nebulous mystery offers

• Trust is worth more than a quick sale.

A buyer who regrets a purchase may never return. A buyer who feels respected may watch every week.

8. Whatnot should be a sales channel, not your entire business.

Whatnot can give sellers access to a large group of active buyers.

It can also change its fees, rules, discovery system, approved categories, or seller requirements.

That is normal for any marketplace. The risk appears when one platform controls nearly every sale your business makes.

A strong seller uses Whatnot to create demand but also builds assets outside the platform, such as

• A recognizable brand

• A website

• An email list built with customer permission

• Social media accounts

• Clear product knowledge

• Reliable suppliers

• Repeatable packing systems

• More than one selling channel

This does not mean sending Whatnot buyers off the platform in ways that break its rules.

It means building a real business that can survive when any single traffic source changes.

The same lesson applies to Amazon, eBay, TikTok Shop, Etsy, Instagram, and almost every large marketplace.

Platforms can provide reach.

They should not own your entire future.

Who should seriously consider Whatnot?

Whatnot may be worth testing when:

• You sell visual, collectible, used, rare, or demonstrable products.

• Your products have enough margin to survive fees and auction risk.

• You can speak clearly about what you sell.

• You have a steady source of inventory.

• You can stream consistently.

• You can pack and ship orders quickly.

• Your country is supported.

• You want to build a community, not only list products.

Whatnot is currently hiring sellers in the United States, Austria, Australia, Belgium, Canada, France, Germany, the Netherlands, and the United Kingdom. Applications will open in July 2026. Sellers from other places, including India, are not eligible at this time. Always verify the official seller location and currency requirements before you invest in a setup.

Who should probably avoid?

Whatnot may not be the right main channel if:

• Your products have tiny profit margins.

• Your products are easy to find anywhere.

• You cannot maintain reliable inventory.

• You do not have time to prepare and ship orders.

• Your products require long research before purchase.

• Your business depends on a calm, premium buying process.

• You expect the platform to provide instant viewers.

• You want passive sales without regular live appearances.

This does not mean your business is weak.

It may simply mean that another sales channel better matches how your customers prefer to buy.

Run a four-show test before making a big commitment.

You do not need to decide whether Whatnot will become your full-time sales channel.

Run a small test first.

Plan four shows using products you can afford to sell. Keep the show length, category, and schedule fairly consistent.

After the fourth show, measure the following:

• Total sales

• Profit after product costs and fees

• Number of items sold

• Average selling price

• Preparation time

• Streaming time

• Packing and shipping time

• Number of repeat buyers

• Mistakes, cancellations, or damaged orders

• Final profit for every hour worked

These numbers will tell you more than a success story on social media.

The final answer: Whatnot is powerful but not simple.

Whatnot can help the right seller turn product knowledge into attention, trust, and sales.

But it does not resolve weak margins, poor inventory, bad shipping, or unwanted products.

The best Whatnot businesses are not simply excellent at livestreaming. They combine entertaining shows with disciplined pricing and reliable operations.

Do not join only because live commerce is growing.

Join when the format matches your products, skills, customers, and available time.

I joined a WhatNot mentorship program called Quick Audience and wrote a detailed review about it.

The real question is not, “Can people make money on Whatnot?”

Clearly, some can.

The better question is, “Can my business make a healthy profit on Whatnot without becoming dependent on it?”

News Summary:

• Google Search Console can now track content from Instagram, TikTok, X, and YouTube.

• The reports display clicks, impressions, search queries, click-through rate, and average position.

• Creators can add each account or channel as a separate Search Console property.

• The reports measure visibility on Google, not views or engagement inside the social platform.

• The feature is rolling out gradually and may not appear in every account yet.

Google Search Console is no longer only for websites.

On July 7, 2026, Google announced platform properties, a new type of Search Console property for Instagram, TikTok, X, and YouTube accounts. The update lets creators and publishers see how their posts perform when people find them through Google Search. Some eligible content can also receive separate reporting for Google Discover and Google News.

What Did Google Add to Search Console?

The platform property connects a social account or video channel to Google Search Console.

Once you verify the account, Search Console shows the Google searches that brought people to its posts. It can also tell you which posts had the most clicks or impressions from Google.

The first supported platforms are:

• Instagram

• TikTok

• X

• YouTube

Each account must be added as its own property. A creator with two Instagram profiles and one YouTube channel would need to set up three separate platform properties.

This builds on an experiment Google introduced in December 2025. That earlier test added some social-channel data to Search Console Insights for website owners. The new system gives social accounts their own properties and can also work for creators who do not have a website.

What Data Will Platform Properties Show?

The main performance report includes the following:

• Total clicks

• Total impressions

• Average click-through rate

• Average search position

• Search queries

• Individual posts that received traffic

Users can filter and sort the data to identify the posts and queries generating the most Google traffic. Google also allows the information to be exported for further analysis.

The Insights report provides a simpler overview of recent traffic, top-performing posts, and the ways people find an account through Google.

There is also an Achievements section. This records milestones such as reaching a new click total during a 28-day period.

The default reporting period is 28 days. New properties may show empty charts at first because Google says it can take several days to collect and process the data. Reporting begins after the property is connected, so users should not expect a complete history from before setup.

Can It Track Google Discover and Google News?

Google says the performance report can include traffic from Search, Discover, and Google News.

But Discover and News reports will only appear when the connected account has received traffic from those surfaces. A creator who has never appeared in Discover will see a non-empty Discover report in the menu.

There is also a reporting detail that may confuse users.

The main insights summary can include clicks from web, image, video, and news searches. However, some of the detailed cards below the main insights summary focus solely on regular web searches. This means the smaller cards may not add up to the total shown at the top.

How to Add an Instagram, TikTok, X, or YouTube Property

Google gives the following setup process:

1. Open Google Search Console.

2. Open the property selector.

3. Select Add property.

4. Choose Instagram, TikTok, X, or YouTube.

5. Follow the authorization steps.

6. Wait a few days for the first reports to appear.

Verification may happen through a direct platform login or through an automated connection with an existing website property. Google will also verify ownership from time to time. If the connection expires or access changes, reporting can pause until the account is verified again.

Adding a property does not improve rankings by itself. It only provides access to performance information.

What Platform Properties Do Not Track

This is the most important limitation.

Platform properties do not replace Instagram Insights, TikTok Analytics, X Analytics, or YouTube Studio. They only show what happens when the account’s content appears on Google.

For example, Search Console will not show the following:

• How many times a TikTok video appeared in TikTok’s feed

• How many Instagram followers viewed a story inside Instagram?

• Native likes, comments, saves, or shares

• Revenue generated inside YouTube

• Reach produced by a platform’s recommendation algorithm

Google does explain how some unusual interactions are counted.

An Instagram Story receives an impression when it appears in Google Search. It receives a click when someone selects it. A video can also receive a Search Console click when it opens inside Google’s own video viewer rather than sending the person directly to YouTube or another platform. (Google Help)

Technical warning: Do not combine Search Console clicks with native platform views and label the total “content reach.” They measure different actions and may include overlapping users.

Why This Matters for Creators

Creators have often had to guess whether Google sends meaningful traffic to their social posts.

A YouTube video might rank for a tutorial query. An Instagram post might appear when someone searches for a product, place, or person. A TikTok video may surface for a how-to question.

Platform properties make more of that behavior visible.

This feature is especially useful for creators who do not own a website. Search Console was traditionally built around verified web properties. Google is now giving social-first publishers access to parts of the same search-performance data. Independent reports from Search Engine Journal and The Verge also highlighted this expansion beyond conventional site owners.

But creators should not treat this expansion as a reason to avoid building an owned platform.

A social account is still controlled by another company. Its reach, rules, and monetization options can change. Search Console data should help creators identify demand they can turn into website articles, email newsletters, products, and landing pages they control.

Marketers should utilize the new query data effectively.

The useful part is not simply seeing more numbers.

The real opportunity is finding search demand that has already been proven by social content.

Use this workflow:

1. Export the queries that produce impressions and clicks.

2. Separate branded searches from non-branded searches.

3. Group the queries by informational, comparison, and buying intent.

4. Identify social posts that rank where your website does not.

5. Turn the strongest topics into detailed owned content.

6. Link the article, video, and social post together where it helps the reader.

7. Review the results after 28 days.

Suppose a short YouTube video receives Google impressions for “how to measure YouTube brand demand.” That may justify creating a longer guide, newsletter, or service page.

OnlineCOSMOS has already covered Google’s new metrics for YouTube brand campaigns, including how advertisers can compare YouTube exposure with later branded searches. Platform properties provide another piece of that measurement process by showing which social and video posts gain visibility in organic Google results.

How Publishers Should Avoid Bad Decisions

Do not rewrite your whole content plan after seeing one high-impression post.

First check:

• Does the query match the content?

• Did it receive clicks or only impressions?

• Is its average position improving?

• Does the topic connect to your products or expertise?

• Could you provide additional insights that enhance the value of the social post?

• Is the traffic likely to continue, or was it tied to short-lived news?

Search Console is evidence, not an automatic instruction.

Publishers should also examine traffic changes alongside Google updates. OnlineCOSMOS’s report on the Google June 2026 spam update explains why changes in clicks and impressions should be reviewed at the query and page level instead of being blamed on one update without proof.

What Is Still Unclear?

Google is rolling out platform properties over several weeks, so some users may not see the option immediately.

The launch currently supports only four platforms. Google has not announced support for Facebook, LinkedIn, Pinterest, Threads, Reddit, or other creator platforms.

It is also too early to determine the amount of useful query data that smaller accounts will receive. Search Console can limit or hide low-volume information for privacy and processing reasons, so the value may differ widely between accounts.

Here are the next steps for creators and marketers.

Open Search Console and review the property selector.

Add every eligible account you control. Then let the reports collect data for at least several days before making decisions.

After the first complete 28-day period, compare:

• Your top social posts in Google

• Your top website pages

• Queries shared by both

• Queries where social content ranks but your website does not

• Topics with clear commercial or newsletter potential

The goal is not to move every social post onto your website.

The goal is to find proven audience questions and build stronger content around the ones that matter to your business.

News Briefs:

• On July 1, 2026, Instagram introduced bilingual captions to its Edits video app.

• The feature automatically translates generated captions into a second language.

• Published reports say it supports 15 languages, including Hindi, Bengali, Gujarati, and Kannada.

• Edits also added stronger templates, clip locking, overlay support, and seasonal sound effects.

Instagram has added bilingual captions to Edits, its separate video-making app.

The feature lets creators place captions in two languages on one video. That may reduce the time needed to prepare Reels for people who speak different languages.

What Did Instagram Add to Edits?

The new tool translates a video’s captions into a second language. This means a creator can show the original caption and a translated version without moving the video into a separate translation app.

Instagram describes the feature as a way to help creators reach new audiences. But the company has not said that using two languages will automatically increase views or improve ranking in the Reels feed.

That difference matters.

Bilingual captions can make a video easier to understand. They do not guarantee more reach.

Which Languages Does the New Caption Tool Support?

India Today reports the same group of 15 supported languages:

• English

• Indonesian

• Russian

• Portuguese

• Gujarati

• Spanish

• Hindi

• Korean

• Bengali

• German

• Italian

• Thai

• French

• Japanese

• Kannada

The inclusion of Hindi, Bengali, Gujarati, and Kannada makes the update especially useful for Indian creators, local businesses, coaches, educators, and regional media pages.

However, the reported launch list does not include Tamil, Telugu, Malayalam, Marathi, Punjabi, and several other widely used Indian languages.

How Do Bilingual Captions Work?

Instagram says edits can automatically translate captions into a secondary language. The company’s short announcement does not provide a full step-by-step help guide.

The likely workflow is straightforward:

1. Update the Edits app.

2. Open or import a video.

3. Generate automatic captions.

4. If you have the choice of bilingual or second language, take that.

5. Choose the translation language.

6. Review both caption versions before export.

Button names may differ by app version or account. This workflow should be checked inside the live app before publishing a detailed tutorial.

Why This Matters for Creators

Language is one of the main limits on how far a video can travel.

A Hindi-speaking creator may have useful content for an English-speaking audience. An English-speaking brand may want to make a product video easier to follow in Hindi, Bengali, Gujarati, or Kannada.

Until now, that often meant copying captions into another tool, translating the text, fixing the timing, and adding it back to the video.

Edits bring more of that work into one app.

For creators who publish several videos each week, even a small reduction in editing time can matter.

Why Marketers and Agencies Should Pay Attention

The feature could make regional campaigns easier to produce.

An agency could create one main video and prepare bilingual versions for different markets. A coach might use English and Hindi. A local retailer might use English and Gujarati. A publisher might use English and Bengali.

But agencies should not treat automatic translation as finished copy.

Names, prices, product terms, local phrases, jokes, and calls to action often suffer from poor translations. Names, prices, product terms, local phrases, jokes, and calls to action can be poorly translated. badly. A wrong word in an advertisement can confuse viewers or change the meaning of an offer.

Expert warning: Translation quality and caption timing are separate problems. A sentence can be translated correctly but still appear too early, too late, or across the wrong scene.

Automatic Translation Still Needs a Human Review

Machine translation (software that changes text from one language to another) can make mistakes.

Creators should pay close attention to:

• People’s names

• Brand and product names

• Prices and dates

• Technical terms

• Local slang

• Formal and informal speech

• Calls to action, such as “buy now” or “book a call”

A native speaker should review important sales, health, legal, financial, or educational content before the video is posted.

For casual content, creators should still watch the full exported video with the sound turned off. This makes it easier to spot poor timing, missing words, or captions that cover important parts of the screen.

Edits Also Added Better Templates and Clip Locking

Bilingual captions are only one part of the update.

Edits now supports richer templates with overlays. An overlay is an item placed above the main video, such as text, an image, a sticker, or another visual layer.

Creators can also lock selected clips inside a template. A locked clip stays in place when someone reuses that template, which can help protect the timing and structure of the original edit.

Instagram also added seasonal sound effects linked to summer content.

These are smaller changes, but they make edits more useful for repeat formats, such as weekly tips and product demos.

• Weekly tips

• Product demos

• News summaries

• Before-and-after videos

• Customer stories

• Educational series

Will Bilingual Captions Increase Reel Views?

Not by themselves.

Captions can help people understand and watch a video without sound. A second language may also make the same video useful to a larger group.

But reach still depends on the topic, opening seconds, viewer interest, watch time, sharing, and how well the video fits its audience.

Instagram has not announced a special ranking boost for videos that use bilingual captions.

Creators should test the feature instead of assuming it works.

A useful test would compare:

• One video with a single language

• A similar video with bilingual captions

• Reach by country

• Average watch time

• Completion rate

• Saves and shares

• Comments in the second language

Run the test across several posts. One successful Reel does not prove that bilingual captions caused the result.

What Should Creators Do Now?

Start with one language pair that matches a real audience.

Do not add a second language only because the feature is new. Choose it because your analytics, comments, customers, or sales data show that people using that language already care about your topic.

Then:

1. Use a short video with clear speech.

2. Generate the original captions.

3. Add the second language.

4. Ask a fluent speaker to check the translation.

5. Keep text away from buttons and lower-screen controls.

6. Compare the post with your normal single-language videos.

The goal is not to place more words on screen.

The goal is to make the video easier for the right people to understand.

What Is Still Unclear?

Instagram’s announcement does not explain:

• Whether the feature is available to every account at the same time

• Whether all 15 language pairs work in both directions

• How the app handles mixed-language speech

• Whether creators can edit each translated word

• Whether translation quality differs by language

• Whether the feature affects caption files when videos are exported

These questions need hands-on testing.

Instagram Edits now gives creators a faster way to place two caption languages on one video.

That can help creators, brands, and agencies prepare content for wider audiences. It may be especially useful in India, where many viewers use more than one language online.

But automatic captions are a starting point—not a finished translation.

Use the tool to save time. Then check the words, timing, and meaning before the video goes live.

Imagine opening Google Maps and finding that the customer reviews you spent years earning are gone.

You may find your five-star rating missing. Your review count may drop to zero. New reviews may stop appearing.

That is what some business owners reported in early July. Google has acknowledged the issue and says it will restore reviews that were removed by mistake, but the company has not given a public recovery date.

Insights Summary:

• Some Google Business Profiles have lost hundreds or thousands of visible reviews.

• Google says suspicious-review detection can remove reviews or temporarily pause new ones.

• The company is investigating whether legitimate reviews were removed incorrectly.

• There is no confirmed timeline for restoring affected reviews.

• Business owners should document the loss before changing their profiles.

1. Some businesses saw years of reviews disappear overnight.

The reports are not limited to one company or industry.

Business owners began posting complaints in the Google Business Profile Help Community, saying large parts of their review history had vanished. One reported profile dropped from about 4,651 reviews to 63. Another said more than 1,200 reviews disappeared, and the profile showed no rating.

Some businesses also reported that new reviews were no longer appearing.

The scale appears to vary. Some profiles lost a smaller number, while others lost nearly every visible review. Google has not said how many businesses are affected.

2. Google says its anti-spam systems may be involved.

Google told Search Engine Roundtable that its systems take action when they find suspicious reviews.

Those actions may include removing reviews and temporarily stopping new reviews from appearing on a profile. Google said the pause aims to prevent further abuse while its systems examine the activity.

“We are investigating the issue and will restore any reviews that were incorrectly removed.”
— Google spokesperson (Search Engine Roundtable)

That wording is important.

Google is not saying that every missing review will return. Reviews that truly break its rules may remain removed. But legitimate reviews caught by mistake could be restored.

Google’s own help pages admit that automated spam detection can sometimes remove genuine customer reviews incorrectly. Businesses are told to contact support when that happens.

3. A fake-review attack may trigger a wider review block.

Several affected businesses said their problems began after their profiles received waves of fake or suspicious reviews.

In some cases, owners reported the fake reviews. Soon after, their legitimate reviews also disappeared or new reviews stopped appearing.

That does not prove that reporting spam caused the loss.

A more likely explanation is that Google’s system detected an unusual review pattern and placed a wider restriction on the profile while it checked the activity. But Google has not publicly confirmed the exact trigger for each affected account.

Its review policy says it may treat unusual volumes or patterns as possible rating manipulation. Google can also temporarily unpublish existing reviews or stop a profile from receiving new ones when it believes fake engagement has occurred.

4. Not every missing review is part of the same incident.

A review can disappear for several normal reasons.

Google says reviews may be delayed while they are checked for policy violations. Reviews can also take time to appear after two business profiles are merged. In some cases, Google temporarily disables user-created content for certain profiles or business categories.

That means owners should first identify the pattern.

A single new review that has not appeared for two days is different from 1,000 reviews disappearing at once.

A sudden, large drop—especially when the rating also falls to zero—is more likely to require support attention than a short delay affecting one customer review.

5. Lost reviews can hurt trust before they hurt rankings.

Review count isn’t just a number.

Often customers use ratings and recent feedback to decide on whether to call, book, visit, or buy. A business that suddenly drops from thousands of reviews to only a few may look new, unpopular, or untrusted—even when nothing about its service has changed.

Reviews may also affect local search visibility. Google says local results are mainly based on relevance, distance, and prominence. Review quantity and positive ratings can contribute to a business’s prominence in local results.

But owners should not assume that missing reviews cause every ranking change.

Local results change for many reasons. The clearest immediate risk is that people who still see the profile will lose customer confidence and lower conversion rates.

This case also shows why businesses should not depend on one platform for all their public proof. Customer stories, testimonials, and case studies should also live on the company’s own website.

OnlineCOSMOS has previously covered how Google’s automated systems can create sudden changes for publishers and businesses, including the Google June 2026 spam update.

6. Document the damage before changing anything.

When reviews disappear, the first thing to do is not panic but document.

Save the following:

1. A screenshot of the current review count and rating.

2. If possible, an older screenshot with the previous count.

3. Date and approximate time changes came up.

4. Business profile name, address & profile URL.

5. Your Google support case number.

6. Links or screenshots from customers whose reviews disappeared.

7. Details on any recent fake review attack or spam reporting activity.

8. Recent changes to the profile, such as a merge, reinstatement, or category edit, are also important.

This record provides Google support with a clearer timeline.

It helps agencies explain the issue to clients as well, without having to rely on memory or partial reports.

7. Do not rebuild the profile or flood it with replacement reviews.

A missing-review incident can make drastic action feel tempting.

Do not create a duplicate business profile to replace the affected one. Do not delete the existing profile. And don’t change the business name, address, or main category unless the information is wrong.

Those changes can create new verification or duplication problems while the review issue is still under investigation.

It is also risky to ask a large group of customers to repost reviews at the same time. Google’s policy treats unusual review volumes and patterns as possible rating manipulation.

A safer approach is to preserve your normal review-request process and wait for clearer guidance before launching a large replacement campaign.

Expert warning: Do not offer discounts, gifts, or payment in exchange for new reviews. Google treats rewarded reviews as fake engagement, even when the customer is real.

8. Use the correct recovery path for the problem you experience.

There are two main routes.

If reviews appear to have vanished because of a technical mistake, use Google Business Profile support and provide the evidence you collected. Google’s help page specifically says businesses should contact support when automated detection removes legitimate reviews.

If Google has sent an email saying your profile received a policy restriction, use the appeal process included in that notice. Google says businesses can appeal restrictions and provide extra context for another review.

Do not report every negative review simply because you dislike it.

Google says reviews are only eligible for removal when they break its policies. A normal complaint from a real customer is not automatically a policy violation.

9. Agencies should tell clients what is known—and what is not.

Local SEO agencies may feel pressure to promise a fast solution

That would be a mistake.

The confirmed facts are limited: Google knows about the issue, suspicious review systems may be involved, the company is investigating, and reviews removed by mistake are expected to return. No public restoration date has been announced.

A clear client message would say:

We have documented the review loss and opened the correct support path. Google has acknowledged a broader issue, but it has not provided a recovery timeline. We are monitoring the profile and avoiding unnecessary edits that could create new problems.

That is more useful than pretending an agency can force Google to restore reviews.

10. The next key signal is whether reviews return automatically.

The most important thing to watch is how Google handles affected profiles.

If review counts return without owners filing individual appeals, that would suggest a broad system correction.

If recovery happens only after support cases, businesses and agencies may need to document and escalate each profile separately.

Google’s public statement leaves both possibilities open.

Until more is known, affected businesses should monitor their profiles in Google Search and Maps, keep screenshots of any changes, and save all support emails.

The larger lesson is simple:

Customer trust should not live in only one company’s database.

Google reviews are valuable. But businesses should also collect customer stories, permission-based testimonials, and proof of results on websites and channels they control.

AI video used to feel like a clever experiment.

You typed a few words, waited, and received a short clip that was sometimes impressive and sometimes deeply strange.

But the market has changed. AI video is becoming a real business, and some of China’s largest technology companies are now putting billions of dollars behind it.

Kuaishou Technology announced on July 2 that investors had agreed to inject about 19.05 billion yuan, equal to roughly $2.8 billion, into the company that will hold its Kling AI business. The investors include companies connected to Alibaba, Tencent, and Baidu, according to Kuaishou’s filing with the Hong Kong Stock Exchange.

News Summary:

• Investors have committed about $2.8 billion to Kling AI.

• The total round could grow to $3 billion.

• Kuaishou placed a $15 billion value on Kling before the new investment.

• Investors include Alibaba, Tencent, and Baidu.

• The deal does not confirm that Kling will be spun off or listed on a stock market.

1. The investor list may matter more than the size of the check.

A $2.8 billion investment is large by almost any measure.

But the more surprising part is who agreed to take part.

The filing identifies an Alibaba Cloud subsidiary, Tencent-controlled companies, a Baidu-controlled business, and a long list of investment funds. That means several major technology groups are backing the same AI video company, even though they compete across cloud computing, advertising, social media, and artificial intelligence.

This is not one parent company quietly funding an internal project.

It is a broad group of investors making a shared bet that AI-generated video will become an important commercial market.

And that matters because the next stage of AI video will require more than clever software. Companies will need huge amounts of computing power, research talent, training data, sales teams, and business partnerships.

Kling now has stronger access to all of those things.

2. Kling is no longer being valued like an experimental AI tool.

Kuaishou said Kling’s pre-transaction value was $15 billion.

“Pre-transaction” means the estimated value of the company before the new investment is added.

When the current $2.8 billion commitment is included, the implied value rises to almost $17.8 billion. That calculation is an estimate based on the figures in the filing, not a separate valuation announced by the company.

This places Kling in a very different category from the small AI tools that launch every week.

Investors are treating it as a large technology business with the chance to sell video generation to consumers, creative teams, and major companies.

Reuters reported that the funding round allows another investor to join within about 60 days. The total amount is capped at roughly 20.45 billion yuan, or $3 billion.

“All eyes will now turn to Kling AI’s upcoming upgrade.”
— Citi analysts, as reported by Reuters

The money creates high expectations.

Kling must now show that it can improve its technology while building a stable and profitable service.

3. Revenue growth helps explain why investors are interested.

The investment is not based only on future promises.

Kuaishou’s filing says the Kling business would have produced about 1.1 billion yuan in revenue during 2025 if the restructuring had already been completed.

It also says Kling reached an annualized revenue run rate of about $500 million in March 2026. An annualized run rate takes revenue from a shorter period and estimates what it would look like across a full year if that pace continued. It is not the same as confirmed yearly revenue.

they make campaign contentReuters separately reported that Kling generated 650 million yuan in the March quarter, more than four times the amount from the same period a year earlier.

Those figures indicate that people and companies are already paying for AI video.

But rapid growth does not automatically mean strong profit. Generating video requires costly computer systems, and AI companies must keep investing to train and operate newer models.

So Kling’s revenue is encouraging, but its long-term costs still matter.

4. This is becoming a fight for advertising—not just viral AI clips.

Many people first saw AI video as a tool for making entertaining clips, cinematic scenes, or social media experiments.

The business goal is much larger.

In its filing, Kuaishou said it expects AI video to spread across:

• Advertising

• E-commerce

• Movies and television

• Short-form dramas

• Animation

• Gaming

For marketers, this technology could change how campaign content is made.

A company might generate several versions of a product video, change the setting for different countries, or test multiple opening scenes without filming each version from scratch.

For creators, it could lower the cost of making visual stories that once needed cameras, actors, locations, and editing teams.

OnlineCOSMOS has already covered how platforms are trying to connect entertainment directly with business results, including TikTok’s Growth Max tools for mini dramas. Kling’s funding suggests the production side of that market is also becoming more competitive.

5. More money does not guarantee better videos.

This is the part investors and headlines can make easy to miss.

Funding can pay for better computers and more researchers. It can help Kling release models faster and reach more customers.

But funding cannot guarantee that every generated video will be accurate, natural, or ready to publish.

AI video systems can still struggle with:

• Faces changing between scenes

• Objects appearing or disappearing

• Unnatural body movement

• Poor text inside videos

• Weak control over detailed prompts

• Inconsistent product shapes and branding

• Unclear commercial-use rules

These problems matter most to businesses.

A strange hand in an experimental art clip may be amusing. A product changing shape inside a paid advertisement can damage trust.

So marketers should judge Kling by usable output—not just model demos, investor names, or funding totals.

6. Creators will have better tools but also faster change.

The funding could help Kling lower generation costs, enhance the quality of the video, and get features out faster.

That would be useful for creators and small businesses that cannot afford traditional video production.

But faster competition creates another problem: tool churn.

A creator may build a full workflow around one AI video platform, only to see another platform release a better or cheaper model a few months later.

That makes long contracts and large prepaid plans more risky.

A safer approach is to keep prompts, scripts, source images, and finished files stored outside any one platform. Businesses should also test the same project across two or three tools before committing to a long-term workflow.

This pattern is not limited to video. OnlineCOSMOS recently examined how lower-cost Chinese models are challenging larger AI companies in its report on GLM-5.2 and the growing budget AI market.

7. Kuaishou is giving up some control—but not the company.

Kling’s assets and operations are being moved into a company called Beijing Kling as part of a wider restructuring.

Kuaishou currently owns the business. But the filing says its stake could fall to about 68.33% if the full investment limit and employee share programs are used.

This is called dilution.

Dilution means a company owns a smaller percentage after new investors or employees receive shares.

Kuaishou would remain the controlling owner under the disclosed plan. But Kling would have more outside shareholders and a more independent structure.

That could make future fundraising easier.

It could also prepare the company for a later spin-off or stock market listing—but neither move has been confirmed.

8. Reports of a future Kling listing should be treated carefully.

Some reports have connected the restructuring to a possible separation from Kuaishou and a future Hong Kong listing.

But the official filing does not announce an initial public offering.

Reuters said Kuaishou had acknowledged early talks about restructuring Kling after earlier media reports of a possible spin-off. The company said those discussions were still at an early stage.

That distinction matters.

A company can restructure a business, bring in investors, and create employee share plans without immediately listing it on a stock exchange.

So the confirmed news is the investment and restructuring.

A Kling IPO remains possible, but it is not yet a settled fact.

9. The real test will be whether Kling becomes dependable.

The funding gives Kling the resources to compete with some of the largest names in AI video.

But creators and marketers do not choose tools because of valuation.

They choose tools because they can produce useful work at a reasonable cost.

Over the coming months, the most important signals will be:

• Whether Kling improves character and product consistency

• Whether generation becomes faster and cheaper

• Whether business-use rights become clearer

• Whether its tools work reliably across languages and markets

• Whether companies can connect Kling to real production systems

• Whether customer revenue grows faster than computing costs

The $2.8 billion investment tells us that major investors believe AI video can become a large business.

It does not tell us who will win.

That will depend on something much less dramatic than a funding headline: which platform can repeatedly turn an idea into a video that a real customer is willing to publish?

A seller turns on a camera, shows some products, initiates an auction, and watches the bids roll in. But the camera is just the visible part.

Behind every smooth Whatnot show, there are good products, clear prices, accurate shipping details, fast packing, and a host who knows how to keep people interested.

Here’s how to start selling live on Whatnot in 2026 without making your first show harder than it needs to be.

1. See if you can sell from your country.

Before investing in lights, inventory, or shipping supplies, verify if Whatnot accepts sellers from your location.

As of July 2026, Whatnot is now accepting seller applications for the following:

• U.S., including Puerto Rico and U.S. Virgin Islands

• Austria

• Australia

• Belgium

• Canada

• France

• Germany

• Netherlands

• The United Kingdom

India and many other countries are not on the approved seller list right now. Your Whatnot account must be linked to a bank account in the same country and currency.

It’s easy to overlook this detail. You can download the app and watch shows, but you may not be able to open a seller account.

“Be certain you qualify before you make a Whatnot business plan.”

2. Apply within the Whatnot app.

Sign up, open the app, and tap “+”. Sell,” then tap “Get Started.”

The application requires:

• Your primary product category

• Your selling experience

• Your regular monthly sales

• Other platforms you’re on

• Your social media profiles

• Your legal name and return address

“Most applicants are granted seller access shortly after they apply, but some accounts may require additional review,” Whatnot says.

Be honest about your experience. You don’t have to pretend you already have a big store.

Some categories, such as luxury bags, wholesale items, and certain surprise formats, require extra approval.

3. Choose one distinct product category.

Don’t make your first show seem like a garage full of random items. Choose one main type of product, for example:

• Vintage clothes

• Sportscards

• Coins, comics, toys

• Jewelry, sneakers

• Cosmetics

Shoppers find it easier to follow a focused show. It also helps Whatnot put your show in the right category, where interested buyers are more likely to see it.

Later, you can work across multiple categories. But when applying, select the category where you expect to sell most often.

A quick test: Complete this sentence—“People should watch my show because I sell ____.” If it takes three sentences to answer, your show is probably too broad.

4. Know your minimum safe selling price.

Live auctions can move quickly. That rush is part of the appeal, but it can also make sellers careless.

Calculate before the show:

• Cost of product

• Platform fees

• Packing costs

• Shipping paid by seller

• Desired profit

There’s an 8% commission fee for many categories in the USA, Canada, and Australia. Payment processing is typically 2.9% of the order total plus a $0.30 transaction fee. Coins and money have a lower commission rate, and Europe and the UK may have different structures, possibly including VAT.

Whatnot does not charge fees to create or store a listing; fees are taken at the point of sale.

For example, Whatnot’s U.S. fee structure shows that an item selling for $50 will net about $44 after the 8% commission and payment-processing fees. The product cost is still deducted from that amount. Time for you!

This is why starting every item at $1 can be risky. A low opening bid can create excitement, but if there are few buyers watching, it can also backfire.

5. Start with a phone, excellent lighting, and stable internet.

You don’t need a television studio. The easiest way to get started is to stream from your phone, says Whatnot. Basic setup requirements:

• A phone with a good camera

• A sturdy phone stand

• Uniform, bright lighting

• A clean table or background

• Good internet connection

• A power bank or charger

Whatnot’s production guide suggests upgrading your phone setup before investing in advanced equipment. Focus on sound and lighting over decorations. Buyers need to hear you and clearly see the item’s color, size, condition, and details. A blurry camera can make a good product look suspicious.

You don’t even have to be on camera. Many sellers just show their hands and products, which can make the first show less stressful.

6. Book the show when you are not quite ready.

After approval, open Seller Hub and schedule your first show. You choose:

• Show title

• Date and time

• Primary category

• Sales format

• Thumbnail image

Whatnot suggests booking about a week in advance. This gives people time to bookmark your show and gives you time to promote it on social media. A compelling title and strong thumbnail can help attract the right audience.

Avoid vague names like “My First Live!” Pick a descriptive title that tells buyers what to expect, like “Vintage Nike Jackets, Tees & Hats for Under $40.” This answers what’s for sale, brands included, and the expected price range.

7. Load your products before you go live.

Don’t leave product organization until the show starts. Pre-set your listings with the following:

• Clear product names

• Good pictures

• Accurate condition descriptions

• Selling method

• Starting or fixed price

• Correct packed weight

By preloading your listings, you can focus on engaging with your viewers, instead of typing product info during the stream. Place your products in the same order as your listings, and number them or store them in labeled containers. This is crucial when you have several sales in a few minutes.

8. Consider your first show to be a rehearsal.

Your initial objective shouldn’t be to become the top seller in your category. Aim for one clean, well-run show.

Here’s a simple plan:

• First 5 minutes: Welcome viewers and introduce your products.

• Next 30-45 minutes: Demonstrate products, answer questions, and hold auctions or fixed-price sales.

• Last 5 minutes: Wrap up and let buyers know when to expect their order to ship.

Whatnot has live auctions, fixed-price listings, and giveaways. Take your time with each product, show all angles and flaws, and describe it before auctioning.

If your audience is small, don’t worry. Ten interested buyers are better than a hundred silent viewers. Some sellers use rehearsal mode (after their first show) to practice pinning products and running test auctions. Private shows are also a great way to test with friends or team members.

9. Remember, the job isn’t over once the livestream ends.

The work isn’t finished when the camera turns off. You still need to:

• Check every order

• Review cancel requests

• Verify package weights

• Print shipping labels

• Pack items securely

• Drop off packages

• Monitor your earnings

Most sellers on Whatnot must ship within two business days of the sale. Packages must be scanned as dropped off or have other approved proof of shipment.

Getting the weights right is important because Whatnot uses them to calculate shipping, select services, and combine orders. The gross weight is the weight of the item plus the packing and packing materials.

New sellers are generally paid after delivery confirmation. When you trigger a payout, funds usually arrive within one or two business days. Experienced sellers may qualify for earlier payouts.

In other words, fast and accurate shipping is part of the sales experience, not just “back office” work.

The real secret isn’t being “good on camera.”

A successful Whatnot seller doesn’t need to sound like a TV host. Be clear, honest, organized, and consistent. Start with a product category, use what you have, set secure pricing, have listings ready before the show, and ship orders quickly.

I’m reviewing Quick Audience profits system by Rachle now, it’s a Whatot mentorship program. I’ll update the evalution report soon in OnlineCOSMOS.

Your first livestream might feel awkward. That’s normal. The better question isn’t, “What if my first show isn’t perfect?”

It’s, “What will I learn by doing the first show that I could never learn by waiting?”

A Google Ads campaign can look healthy while hiding a costly problem.

Maybe you told Google that you were willing to pay $10 for each lead. But the campaign has recently brought in leads for only $5 each.

That sounds like great news. And it is—until Google starts taking the old $10 target more seriously.

Starting August 17, 2026, Google will change how some budget-limited campaigns use target-based bidding. The update is meant to make campaign results more predictable. But advertisers who have not reviewed their targets could see their cost per lead rise or their return on ad spend fall, according to Google’s official notice about the bidding change.

News Summary:

• The change starts on August 17, 2026.

• It mainly affects campaigns marked “Limited by budget.”

• Campaigns beating their Target CPA or Target ROAS may move closer to those stated targets.

• Google says it will not automatically change an advertiser’s target or daily budget.

1. Google is not raising your budget—it is following your target more closely.

This is the first important distinction.

Google is not announcing an automatic budget increase. Daily and monthly spending limits will still apply.

Instead, the bidding system will follow the performance target you entered more closely when the campaign is limited by budget, as explained in Google’s target-based bidding FAQ.

A target CPA is the average amount you want to spend on a single conversion, such as a sale, signup, or lead. A target ROAS is the revenue you want to achieve divided by what you spend on ads.

For example, imagine that your Target CPA is $10, but your campaign has recently produced conversions for $5 each. Google says the campaign may begin delivering closer to the $10 target after the update unless you lower the target.

“Google won’t automatically adjust your bidding targets or budgets.”
— Google Ads Help

That sentence matters. Google is giving advertisers more predictable control. But it is also placing more responsibility on them to enter the right numbers.

2. The campaigns doing best today may face the most significant change.

Strong campaigns can also feel the update.

They may be the campaigns that are performing much better than the target you originally set.

Search Engine Land’s report on the update notes that budget-limited campaigns that are outperforming their Target CPA or Target ROAS may no longer do so automatically after August 17. Results could move closer to the configured target instead of recent real-world performance.

That creates a strange situation. A campaign can have an old target that once made sense but no longer reflects the business.

Better ads, improved landing pages, stronger brand awareness, or cleaner conversion tracking may have reduced the real CPA over time. But Google does not know whether the old target is outdated.

It only knows the target you set.

3. “Limited by budget” is the warning label to watch.

The update does not affect every Google Ads campaign.

Google says that Target CPA and Target ROAS campaigns that are not constrained by budget will continue to operate as they do now. Manual CPC and Target Impression Share are also not affected by this specific change.

The affected group includes budget-limited campaigns using Target CPA, Target ROAS, and Target CPC for Demand Gen. The update covers Search, Shopping, Performance Max, Demand Gen, Travel, and other supported campaign types listed in Google’s official FAQ.

A campaign marked “Limited by budget” is one that Google believes could receive more useful traffic if it had a larger budget.

But the label does not mean you should blindly raise the budget. First First, assess whether the campaign is profitable, whether its conversion data is accurate, and whether its current target still matches the business goal.

4. Performance Max and Demand Gen may shift where the money goes.

The change may be harder to read inside campaigns that use several Google surfaces.

Performance Max and Demand Gen can place ads across different channels. Google says the overall bidding change will work in the same way, but advertisers may notice changes in how spending is divided between those channels.

That does not automatically mean the campaign is broken. It means advertisers should save a clear baseline before August 17: channel mix, conversion volume, CPA, ROAS, lead quality, and total spend.

This update also arrives as Google moves more advertising work into automated systems. OnlineCOSMOS has already covered Google Display Ads moving into Demand Gen and the new YouTube brand campaign metrics. Together, these changes make account-level measurement more important, not less.

5. Advertisers have four real choices.

Google’s guidance does not say that every advertiser must lower a target.

There are four main choices:

• Keep the current target. This makes sense when the number still reflects what the business is truly willing to pay.

• Update the target to match recent performance. If you want to preserve that level of efficiency, a campaign that is delivering $5 conversions at a $10 target CPA could be pushed down to a $5 target.

• Choose a new business-based target. Recent performance is not always the right goal. A $7 target may leave enough profit but still allow for more growth.

• Switch to a volume-based strategy. Using “Maximize Conversions” or “Maximize Conversion Value” may bring more results for your budget, but the CPA or ROAS may vary more when you adjust your budget.

The correct choice depends on profit—not on which number looks best inside Google Ads.

A very low CPA is not useful if the leads are poor. And a high ROAS is not enough if the campaign produces too little revenue to matter.

6. The audit should happen before the rollout, not after.

Google’s Bid Target Adjustment Tool is scheduled to begin appearing in accounts on July 6, 2026. It will show historical performance and help advertisers review or update affected targets, according to Google’s rollout guidance.

Before accepting any recommendation, review these five items:

1. Identify campaigns marked “Limited by budget.”

2. Compare your actual results over the last 30 to 90 days to the Target CPA or Target ROAS you set.

3. Review whether the conversions represent real sales or qualified leads.

4. Calculate the highest CPA—or lowest ROAS—the business can accept while remaining profitable.

5. Save the current budgets, targets, results, and channel allocation before making changes.

Don’t change every campaign at once without a good reason.

When you change a target or budget, Google recommends waiting one or two conversion cycles to see the results. A conversion cycle is the normal time between an ad click and the completed sale or lead.

7. Forecasts may be less reliable during the first two weeks.

Expert flag: Google says advertisers should be extra cautious with their forecasts from August 17 to August 31, as its planning systems adjust to the new bidding behavior.

That detail could be easy to miss.

Businesses planning large promotions during the second half of August should not treat forecast numbers as guaranteed outcomes. Use them as estimates and compare them with actual conversion data, profit margins, and past seasonal results.

8. An outdated target can become a costly business instruction.

The update sounds technical, but the risk is simple.

An old number inside Google Ads may become a stronger instruction after August 17.

For a small business, that could mean paying more for a lead without quickly understanding why. For an agency, it could create a client problem: a campaign may appear to lose efficiency even though Google is simply moving it closer to the goal already stored in the account.

So the key question is not, “What CPA has Google been giving us?”

It is, “What CPA can this business afford while still making a healthy profit?”

9. The bigger lesson: automation follows the numbers you give it.

Google plans to start offering the adjustment tool on July 6, followed by the bidding change on August 17. Advertisers may then need several conversion cycles to understand the full effect.

The wider lesson goes beyond this one update.

Automated ad systems work from the goals, values, and conversion data that people provide. When those inputs are old or wrong, more automation can make the mistake larger—not smaller.

Before August 17, open your Google Ads account and check one simple gap:

Is the target you entered still the result your business actually wants?

.

.

.

Topics covered: Google Ads, Target CPA, Target ROAS, Performance Max, Demand Gen, Search Ads 360, Display, and Video 360.

Insights Summary:

• Microsoft sends 6,000 experts and $2.5 billion to fix busted enterprise AI

• • The UN forms a new commission to establish global AI rules, and the US quietly negotiates voluntary release standards.

• GitHub demonstrated that remediating code leaks is more difficult than detecting them, clearing a massive backlog of 20,000 alerts.

• Ad agency leaders warn AI is for workflow speed, not taking over your brand’s creative soul.

Are your AI pilots making money or just burning API credits? Microsoft’s newest $2.5 billion bet suggests most businesses are still on the runway.

The Death of the AI Sandbox

So you bought access to AI. What next?

Microsoft has come to realize it needs to do more than just sell AI models. The Official Microsoft Blog reports that they launched the Microsoft Frontier company, investing $2.5 billion and sending 6,000 specialists directly into customers’ businesses.

But why the big switch? Because it is incredibly challenging to plug an LLM (a large language model, or smart text generator) into messy company data.

Microsoft’s move highlights that the real bottleneck is implementation, the actual setup work. Now, the real competitive advantage is in the execution. “If you’re just selling basic AI consulting and don’t have real workflow skills, you are in danger of being squeezed out.

The Game’s Rules Are Changing

Governments scramble to set boundaries, while tech giants build.

The International Telecommunication Union (ITU) has just launched the AI for Good Global Commission. They want to push global AI collaboration.

But behind closed doors in the US progress is being made. The government is discussing voluntary standards for the release of new AI models, Reuters reports.

Technical Alert: Please note this US news is unconfirmed. Don’t build your product roadmap on rumors of non-existent laws. Distinguish between binding legislation (laws you must comply with) and voluntary guidelines.

The Ugly Truth of Code Security

It’s easy to find a security hole. The real nightmare is fixing it. GitHub revealed it had to deal with more than 20,000 secret-scanning alerts in 15,000 repositories (folders to store code).

They didn’t just erase old passwords. They built a huge system to identify owners and when to revoke access altogether. If your team is all-in on AI, your API keys (the digital VIP passes to your software) are probably leaking out as we speak. Fix the problem by setting up pre-commit scanning today to catch leaks before they happen.

The brand is still owned by humans.

All this technology, and AI still can’t run your marketing department.

Digitas CEO Amy Lanzi told The Verge in a recent interview, “AI is great for workflow speed. AI can spit out variations of ad copy all day.

But it’s dangerous to leave creative control entirely to an automated system. It damages long-term brand building. There’s still a need for human judgment to defend what your company stands for.

The reason your day-to-day will change

This isn’t just news for tech billionaires. It also impacts your daily workflow.

If you run a digital business, soon your clients will demand proof – they’ll want to know exactly how your AI works, with audit trails and security assurances. Gone are the days when you could just add a basic AI chatbot to your website and call it a day.

So what should you do today?

Don’t try all the new AI toys that come along. Today, audit your current workflows. What is a concrete task that AI can deliver tangible business outcomes from? Keep your API keys safe. Don’t let an algorithm define your brand’s main message.

.

.

.

Topics Covered: Microsoft, Microsoft Frontier Company, UN International Telecommunication Union, GitHub, Digitas

• Claude Fable 5 became globally available again on July 1, 2026, after U.S. export restrictions (government rules about selling technology to other countries) were lifted on June 30.

• The more restricted Mythos 5 model is only available to a specific set of U.S. organizations with government approval.

• API pricing is $10 per million input tokens and $50 per million output tokens, with a 90% discount available through prompt caching (saving part of the input so the AI doesn’t re-read it).

• Anthropic added new cybersecurity classifiers (safety filters) that can automatically reroute flagged requests to Claude Opus 4.8, a less powerful model.

Anthropic’s most capable public AI model is back online—but it returned with new rules. Claude Fable 5 went live worldwide on July 1 after the U.S. Department of Commerce dropped the export controls that had shut it down on June 12. The model is powerful. It’s also expensive, newly filtered, and subject to data-retention policies that your compliance team needs to read before you deploy it.

What Happened: The Export Ban and the Lift

On June 12, 2026, the U.S. government issued an export control directive, an official order that restricted the technology it could share with other countries. That order forced Anthropic to suspend access to both Claude Fable 5 and the more advanced Claude Mythos 5 for users outside approved channels.

The restriction lasted about two and a half weeks. On June 30, the U.S. Department of Commerce lifted the directive. Anthropic began restoring Fable 5 access the very next day, July 1.

But the two models didn’t come back equally. Fable 5 is globally available. Mythos 5 is available only to a specific set of U.S. organizations that received government approval. It is stated that Anthropic is collaborating with the government to expand that list.

This is important as it sets a precedent (a model that future decisions may follow). A government can turn off an AI model for millions of users in a single day. And when it comes back, it may come back with restrictions that didn’t exist before.

How Much Claude Fable 5 Costs and What You Actually Pay

The published API price is straightforward: $10 per million input tokens and $50 per million output tokens. That makes Fable 5 exactly twice the cost of Claude Opus 4.8, which sits at $5/$25.

Anthropic offers a 90% discount on input tokens through prompt caching. If you’re running repetitive tasks with a large shared context—like analyzing the same document library over and over—caching can bring your effective input cost down to roughly $1 per million tokens.

There’s a catch for U.S.-specific workloads. If your inference (the process of getting an answer from the model) must happen in the United States, Anthropic applies a 1.1x pricing multiplier on both input and output tokens. That bumps your effective rate to $11 input / $55 output per million tokens.

The model is also available through cloud providers like AWS (Amazon Bedrock), Google Cloud, and Microsoft Foundry, each of which may add its markup or offer its own volume discounts.

Through July 7, 2026, Fable 5 usage is capped at 50% of weekly usage limits for Pro, Max, Team, and select Enterprise plans. After that date, usage transitions to a credit-based system. Anthropic hasn’t published the full credit-pricing details yet.

The Data Retention Policy You Can’t Afford to Skip

Here’s the detail that matters more than any benchmark score.

Anthropic’s documentation states that certain Mythos-class traffic (requests handled by the most powerful models) can carry a 30-day data retention period. That means Anthropic may store the data you send through the model for up to 30 days.

For a solo developer testing code snippets, that’s probably fine. For a company handling customer records, financial data, legal documents, health information, or proprietary business logic, 30-day retention creates a compliance question that requires a legal review.

The key questions your team should ask before deployment:

1. What data is retained? All input/output, or only flagged requests?

2. Where is it stored? Which data center, which country, and which jurisdiction (whose laws apply)?

3. Who can access it? Anthropic staff, government agencies with subpoenas, or nobody?

4. Can you opt out? Is there a zero-retention API tier for enterprise customers?

5. Does retention differ between Fable and Mythos? Anthropic documents different conditions for each.

If you are unable to answer these questions from the official documentation, please refrain from deploying the model on sensitive workloads until you have the necessary information.

The Safety Filter That Can Silently Change Your Output

Anthropic redeployed Fable 5 with enhanced cybersecurity classifiers—safety filters designed to block requests that the system judges as potentially harmful, particularly in cybersecurity-related domains.

That sounds reasonable in principle. But here’s the operational risk most people miss.

If your prompt triggers the new classifier, the system doesn’t just block your request. It can automatically reroute your request to Claude Opus 4.8—a less powerful, less expensive model. You may not realize the rerouting happened unless you’re checking which model actually processed your query.

That means your output quality and your cost can both change in the middle of a workflow, without warning. If you’re running batch processing (sending many requests in a row), you could end up with a mix of Fable-quality and Opus-quality outputs, and you’d only know by inspecting the response metadata.

⚠️ Expert Flag: Before going live with Fable 5, run your full prompt library through the model and check which requests trigger the fallback. Document the patterns. If you’re building a product on top of Claude, your users will notice inconsistent quality—even if you don’t.

How Fable 5’s Return Changes Your AI Tool Choices

Here’s what this decision means if you’re not an AI engineer.

If you’re a marketer or agency operator, Fable 5’s return gives you access to one of the most capable AI research and writing models available. You can use it for deep competitive analysis, long-form content drafts, data synthesis (combining information from many sources into one summary), and multi-step campaign planning. But your clients’ data policies need to allow 30-day retention. Ask before you paste their proprietary information into a prompt.

If you’re a startup founder, more model choice sounds fantastic—until you see the bill. At $50 per million output tokens, a careless deployment that generates many long responses can burn through budget fast. The 90% caching discount helps, but only if your workload structure supports it. Use Fable 5 for the hard tasks. Route the easy ones to cheaper models like GLM-5.2 or Opus 4.8.

If you run affiliate or growth content, this is a high-search-volume moment. People are searching for “Claude Fable 5 pricing,” “Claude Fable 5 vs GPT,” “is Claude Fable 5 available,” and “Claude data retention policy.” “That’s four distinct content pieces from one news event. (Editorial analysis)

Key Dates and Decisions to Watch After July 7

Three things will shape this story over the next month.

The credit-based pricing transition. After July 7, Anthropic shifts Fable 5 from percentage-based usage limits to a credit system. The details of that system—how credits are earned, purchased, or allocated—will determine whether Fable 5 remains affordable for mid-tier teams or becomes a premium tool only large enterprises can sustain.

Mythos 5 access expansion. Anthropic says it’s working with the U.S. government to expand Mythos 5 access beyond the current approved list. If Mythos becomes broadly available, it could shift the competitive picture again. But the political uncertainty around AI export controls means that timeline is unpredictable.

Competitor responses. GLM-5.2 is gaining ground at the budget end. Google’s Gemini models are competing at the premium end. Anthropic’s ability to hold its position depends on whether Fable 5 delivers measurably better results on high-stakes tasks—not just benchmark scores, but real production outcomes that justify the 5–10x price premium.

Fable 5 Is Available—But “Available” Isn’t the Whole Story

Claude Fable 5 is back, and that’s welcome news for the AI ecosystem. More model choice means more competition, which means better tools and eventually lower prices for everyone.

But this episode also showed something uncomfortable: a government directive can take a globally used AI model offline in a single day. And when it comes back, it comes back with new filters, new retention rules, and new pricing structures that didn’t exist before.

The lesson isn’t “don’t use Fable 5.” The lesson is “don’t build your entire product on a single model from a single provider, subject to a single government’s export policies.” Diversify your model stack the same way you’d diversify any critical supply chain.

Your next step: build a compliance comparison matrix for Fable 5 vs. your current model. Cover access, retention, safety filters, fallback behavior, pricing (including the U.S. multiplier and caching discounts), and export-control risk. Then make your decision based on your specific compliance environment—not on a headline.

Why Developers Are Switching to GLM-5.2

Three words: cost, context, and openness.

GLM-5.2 is MIT-licensed. That means its weights are free to download and use. It handles a 1-million-token context window. That’s enough to feed it an entire codebase (a collection of all the code in a software project) at once. And it was built for long-horizon (long-running, multi-step) coding tasks and agent workflows (programs that make decisions on their own).

According to Z.ai‘s documentation, the model uses a technique called “IndexShare” to cut computing costs during long-context work.

The pricing isn’t one number; here’s why that matters.

Don’t trust a single cost figure for GLM-5.2. Prices change based on where you buy access.

On OpenRouter, rates range from roughly $0.93 to $1.40 per million input tokens and $3.00 to $4.40 per million output tokens, depending on the hosting provider. Cached-read pricing drops to around $0.18–$0.26 per million tokens for repeat workloads.

Compare that to Claude Fable 5’s $10/M input and $50/M output, and the gap is massive—on paper.

But here’s the flag most coverage misses: token price isn’t task price. A model that fails 30% of coding jobs will cost more per successful job than a model that fails 5%, even if its per-token rate is 5X higher. You need to measure the cost per successful task—including retries, developer oversight, and latency.

⚠️ Technical Warning: GLM-5.2 was reportedly trained entirely on Huawei Ascend chips. For companies with strict vendor-origin policies or government contracts, such a situation creates a compliance (following rules) question that has nothing to do with performance. Please verify your procurement rules before routing production traffic.

What This Means for Indian Startups and Global Teams

Indian startups are able to cut API costs in dollar terms by sending less demanding tasks to GLM-5.2 and reserving premium models for more difficult jobs.

But data governance (rules governing where data goes and who can see it) adds friction. If your customer data passes through a Chinese-origin model at a third-party provider, you will want to get clear answers on storage, logging, and jurisdiction (whose laws apply).

So what do you do? Run 20 identical production tasks through GLM-5.2 and your current model. Track completion rates, latency, retry costs, and total developer time. That real-world number—not the pricing page—tells you whether switching saves money.