Demur Design white logo icon: branding symbol
Back to all insights
personalized pricing FTC warning daily roundup for small business

Personalized Pricing Draws an FTC Warning, and AI Max Goes Global

On August 19, 2026, the Federal Trade Commission put every business that sets prices with customer data on notice. The agency voted to seek public comment on a draft enforcement policy statement saying that undisclosed personalized pricing, the practice of using someone’s personal data to decide what they should pay, can break federal law. It is not a rule and it is not a ban. It is a written warning from the regulator that enforces unfair and deceptive practices, and the comment window is short.

Most owners hear “surveillance pricing” or “personalized pricing” and picture an airline or a national retailer. That is the wrong picture. If you run an online store, a booking system, a quoting tool, or any plugin that adjusts what a visitor sees based on who they appear to be, this statement is aimed at you too.

Two other changes landed in the same window: Microsoft turned its AI Max automation on for advertisers worldwide, and Google began pushing generated, interactive layouts into AI Overviews. Here is what each one means and what to do about it.

The FTC Just Warned Businesses About Personalized Pricing

The short version: the FTC believes that hiding the fact that you use personalized pricing is a deceptive practice, and it intends to enforce on that basis. The Commission vote authorizing the notice was 2-0, and once the statement is published in the Federal Register the public gets 30 days to comment.

The agency was careful about the limits of its own authority. In its announcement it said plainly that it “does not have the legal authority to ban personalized pricing in all circumstances,” but that “businesses that fail to tell consumers how their personal data is being used to set a price may be in violation of the FTC Act and other laws we enforce.” The problem is not the pricing. The problem is the silence around it.

What the statement actually says

The draft rests on a simple expectation about how shoppers read a price tag. When a customer sees a listed number, they assume everyone else sees the same number. They expect prices to move with supply and demand, not with their browsing history or their past orders. That assumption is exactly what personalized pricing quietly breaks.

Under that reasoning, a seller who represents or implies that a price is fixed when it actually varies by individual is at risk of misleading the customer. The FTC also flags the collection side. Gathering or using personal data specifically to power personalized pricing, without telling anyone, could itself violate Section 5 of the FTC Act, which prohibits unfair or deceptive acts in commerce.

The consumer-harm argument the agency makes is worth understanding, because it explains why disclosure is the fix. An informed shopper can respond to personalized pricing. They can switch to a private browsing session, use a VPN, or shop somewhere else entirely. A shopper who does not know it is happening has no way to protect themselves, and that information gap is what turns an ordinary business decision into a legal exposure.

What counts, and what does not

Ordinary commercial pricing is not the target. Seasonal changes, published sale periods, quantity discounts, member rates that anyone can sign up for, and rates that move with demand across all customers are all normal and visible. None of that is personalized pricing in the sense the FTC is describing.

What the agency is describing is narrower and quieter: a number calculated for one individual, from data about that individual, that a different visitor would never see. Location signals, device type, browsing history, past purchase behavior, and inferred household details are all inputs that can produce it. The examples in the coverage are blunt, including a delivery service charging more for milk because household data suggests children live there, or a hotel raising a rate because the data says the visitor is already traveling.

Three things to check this week

First, find out whether you are using personalized pricing at all. Many owners are not aware they are, because the behavior arrives inside software. Ecommerce personalization apps, dynamic pricing plugins, AI quoting tools, and some booking platforms adjust displayed rates using visitor signals by default. Open the settings and read what the feature actually does rather than what the marketing copy says.

Second, if you are doing it, say so. A plain line on the pricing page and a matching line in the privacy policy explaining that prices may vary based on information collected about the visitor is not complicated to write. That one disclosure is what moves personalized pricing from a quiet legal exposure to an ordinary business practice, and it is the simplest form of compliance available.

Third, check your claims. If your site says “our price” or shows a single figure in a way that implies it is the same for everyone, that language has to match reality. This is the same discipline that already applies to fees and add-ons, and it is where most exposure to a personalized pricing complaint would come from.

One more note on timing. Because the personalized pricing statement is a draft open for comment, the terms can still change, and a trade association or a software vendor you rely on may file something worth reading. Watching the docket for 30 days is free.

Microsoft Turned AI Max On for Advertisers Everywhere

Microsoft Advertising began rolling out AI Max to all accounts globally, moving the feature out of the open pilot it started in May and into general availability. If you run Search campaigns on Microsoft, the AI Max toggle should now appear in your account.

AI Max bundles three behaviors, and it helps to understand them separately rather than as one switch. Search term matching lets a campaign enter auctions for queries beyond your keyword list, using signals from your keywords, ads, and landing pages. Text customization generates additional ad copy from your existing assets and site content. Final URL expansion can send a click to a different page on your site when the system decides that page fits the query better.

Microsoft describes the point of all three as reaching more conversational searches, including the ones happening inside Bing and Copilot. That is the real reason this exists. Query language is getting longer and messier as people talk to search engines, and a fixed keyword list catches less of it every quarter.

The controls shipped alongside it, which is the detail worth crediting. Brand inclusions and exclusions shape which branded queries you show on, term exclusions keep specific phrases out of generated copy, URL rules limit where expansion is allowed to send traffic, and the whole thing can be applied at the ad group level rather than all at once. Search term and asset reporting were there from the start, so you can see what the automation actually did.

The practical advice for a small advertiser is the same as it was for every automation rollout before this one. Turn it on in one ad group, not the account. Set brand and term exclusions before you flip anything, because generated copy is far easier to prevent than to retract. Then read the search term report after a week and decide with data. If your budget is small, uncontrolled query expansion spends it fastest, which is the same trap we covered when Google changed how its bidding targets work.

Google Is Building Interactive Answers Into AI Overviews

Google confirmed that generative UI, the capability that assembles custom visual layouts and interactive tools on the fly, is now live globally in English in AI Mode and has, in Google’s words, “started rolling out as well in AI Overviews.” Full availability there is expected over the coming weeks.

It arrived inside a back-to-school announcement about study tools, alongside practice quizzes and step-by-step help in Lens, so it is easy to file as an education story and move on. That would be a mistake. AI Overviews sit on the same results pages your customers use, and the change makes those results richer and more self-contained.

For a small business, the consequence is the one we have been tracking all year. When the answer becomes an interactive thing a searcher can use on the results page, the reason to click a website gets thinner. The response is not to fight it. It is to make sure the pages that do earn clicks are the ones tied to a decision or a transaction, because informational pages are where this trend bites hardest. Our breakdown of how search, local rankings, and AI answers fit together walks through that split in detail.

Personalized Pricing and This Week’s Changes: Common Questions

Is personalized pricing illegal now?

No. The FTC said directly that it cannot ban the practice outright. What the draft statement targets is doing it without telling anyone. The agency’s position is that undisclosed personalized pricing may be a deceptive practice under the FTC Act, which makes disclosure the deciding factor.

Does a sale or a member discount count?

Generally no. Published sales, seasonal rates, quantity breaks, and loyalty pricing that any customer can qualify for are visible and available to everyone. The personalized pricing concern is an individualized number built from personal data that a different visitor would never be shown, and that the shopper has no way to detect.

How would I know if my website is doing this?

Check your tools rather than your intentions. Ecommerce personalization apps, dynamic pricing plugins, and AI quoting features often introduce personalized pricing from visitor signals by default. Open each setting and confirm what inputs it uses. If a vendor cannot answer that question clearly, treat the answer as yes.

Should I turn AI Max on right away?

Turn it on in a single ad group first. Set your brand exclusions and term exclusions before enabling anything, since generated copy is easier to prevent than to walk back. Give it about a week, read the search term report, and expand only where the added queries were genuinely relevant.

Will interactive AI Overviews cut my traffic?

For informational queries, most likely yes over time. Pages that answer a simple question are the easiest for a generated answer to replace. Pages tied to a purchase, a booking, a quote, or a local visit hold up better, which is where your effort belongs while this rolls out.

What to Take From This Week

Three changes, one theme: software is quietly making decisions on your behalf, and you are accountable for them. Personalized pricing is the clearest case, because a plugin you never audited can create a disclosure obligation you did not know you had. AI Max makes the same point in advertising, and generated search answers make it in organic search. In all three, the owners who come out ahead are the ones who open the settings and read them.

Stay Ahead of Changes Like These

Search results, ad platforms, and pricing rules are all moving at once, and the same shift is showing up in how AI tools, search, and laws now overlap for small businesses. If you want your site audited for what it claims and what it actually does, our SEO and analytics service covers exactly that ground. If you are unsure whether a plugin on your site is running personalized pricing from visitor data, get in touch and we will look at it with you. To catch the next change the day it happens, subscribe to the Demur Design newsletter in the footer below.

This recap is researched and drafted with AI, then reviewed, fact-checked, and published by Demur Design.

Sources

more insights

Demur Design white logo icon: branding symbol