TL;DR
- The FTC announced on 19 August 2026 that it is seeking public comment on a proposed enforcement policy statement on personalised pricing, defined as using personal data to set prices according to what a company believes an individual consumer is willing to spend.
- The proposed statement says failing to disclose that a price has been personalised, the basis for the personalisation and the types of data behind it, is likely to constitute an unfair or deceptive act or practice in violation of Section 5.
- The Commission vote authorising the Federal Register notice was 2-0, and the public will have 30 days to comment once the notice publishes.
- The FTC states it lacks authority to ban personalised pricing outright, and the statement carries a disclaimer that it confers no rights and does not bind the Commission or the public.
- The statement distinguishes ordinary supply-and-demand variation, regional and tax differences, and inherently individualised products such as insurance premiums and credit terms, from individualised pricing on goods where shoppers expect one listed price.
The Federal Trade Commission announced on 19 August 2026 that it is seeking public comment on a proposed enforcement policy statement on personalised pricing, which it defines as using personal data to set prices according to what a company believes an individual consumer is willing to spend. The proposed statement says that failing to disclose that a price has been personalised, along with the basis for the personalisation and the types of data behind it, "is likely to constitute an unfair or deceptive act or practice in violation of Section 5." The Commission vote authorising the Federal Register notice was 2-0, and the public will have 30 days to comment once the notice publishes.
What the FTC actually proposed on 19 August 2026
The document is a proposed enforcement policy statement, not a rule. It signals how the Commission intends to read existing Section 5 authority, and the FTC attached a disclaimer stating that it confers no rights and does not bind the Commission or the public. The FTC also stated that it lacks authority to ban personalised pricing outright. Read together, those two facts define the shape of the thing: the FTC is not outlawing the practice, it is putting a disclosure obligation around it and telling the market how enforcement will likely be framed.
FTC Chairman Andrew Ferguson framed the consumer expectation directly in the announcement: "When consumers see a listed price, they expect it to be same price that everyone else sees, not the retailer's estimate of how much they are willing to pay based on their personal data."
The Commission published the announcement on ftc.gov, and the press release is available in full there.
The proposal in the FTC's own terms
The announcement carries a small number of load-bearing facts, and several of them are procedural rather than substantive. Separating them keeps the status of the document clear.
| Element | What the FTC stated on 19 August 2026 | Status |
|---|---|---|
| Definition of personalised pricing | Using personal data to set prices according to what a company believes an individual consumer is willing to spend | Proposed policy statement, out for comment |
| The disclosure standard | Failing to disclose that a price has been personalised, the basis for the personalisation and the types of data behind it, "is likely to constitute an unfair or deceptive act or practice in violation of Section 5" | Proposed, binds nobody |
| Commission vote | 2-0 to authorise the Federal Register notice | Taken |
| Comment period | 30 days for the public to comment | Runs once the Federal Register notice publishes |
| The FTC's stated limits | The Commission lacks authority to ban personalised pricing outright, and the statement confers no rights and does not bind the Commission or the public | Stated in the proposal itself |
The line being drawn is between what you want and what you will pay
Personalisation as most marketers practise it answers the question of what a given person wants to see. Which products surface first, which creative they get, which email lands, which category page sorts to the top. Nothing in the FTC's 19 August announcement disturbs any of that.
Personalised pricing, as the FTC defines it, answers a different question: what will this person pay. The input is still personal data, but the output is the number on the price tag rather than the ordering of the shelf. The Commission's position in the proposed statement is that a shopper looking at a listed price carries an expectation that the number is the same one everyone else sees, and that quietly making it a per-person number without saying so is where Section 5 exposure begins.
That is a narrower target than "personalisation is under scrutiny" headlines will suggest. The obligation the FTC proposes is disclosure, and it has three named parts: that the price was personalised, the basis on which it was personalised, and the types of data used. A company that personalises price and says so on the page is addressing what the proposed statement asks for. A company that does it silently is the one the statement is aimed at. The target throughout is undisclosed personalised pricing rather than personalisation in general.
Three things marketers routinely blur
Most of the confusion around this announcement comes from collapsing three separate practices into one word. The FTC's definition only reaches the third. What follows is analysis, not FTC guidance.
- Dynamic pricing. The price moves with conditions that are not about the individual: time of day, seasonality, inventory, demand. Everyone hitting the page in that moment sees the same number. The FTC's proposed statement expressly distinguishes ordinary supply-and-demand variation from what it is targeting.
- Personalised promotion. A member coupon, a tier discount, a welcome-back offer. The list price is unchanged and identical for everyone; a named, disclosed offer is applied on top of it for a person who qualifies. The shopper can see the offer, see the mechanism and see the original price.
- Personalised pricing. The listed price itself is computed from what the company estimates this individual will spend. There is no visible offer, no stated qualification and no reference price. This is the practice the FTC's proposed statement addresses.
The practical test that falls out of this is whether a shopper could describe why their number differs from someone else's. With a promotion they can: they joined the loyalty programme, they hit gold tier, they used a code. With personalised pricing they cannot, because the mechanism is the estimate itself.
Why a loyalty discount sits on the safe side, if it is built as a promotion
The structural difference is where the personalisation attaches. A promotion attaches to an offer: there is a base price on the page, a named discount, and a stated reason the shopper qualifies for it. A personalised price attaches to the person: the base price is the personalised number and there is nothing behind it to compare against.
That distinction is worth auditing rather than assuming, because plenty of loyalty stacks drift across it without anyone deciding to. A pricing engine that starts by suppressing the reference price for high-propensity segments, or one that computes a member price directly rather than applying a visible discount to a list price, has quietly moved from the second category to the third. The commercial outcome may be identical. The disclosure posture is not.
Anyone running this in an ecommerce programme can check it on the product page itself. Is there a list price visible? Is the discount named? Could a shopper explain the reason they got it? Three yeses put the mechanism on the promotion side of the line the FTC drew.
What the FTC placed outside the definition
The proposed statement distinguishes several things from the practice it targets. Ordinary supply-and-demand variation is outside it. Regional and tax differences are outside it. So are inherently individualised products, and the FTC named insurance premiums and credit terms as examples. Those are categories where the shopper already understands that the number is computed for them, and where individualised pricing is the product rather than a hidden layer on top of it.
What the statement targets is applying individualised pricing logic to goods where shoppers expect a single shelf or listing price. That framing does real work. It means the exposure is highest exactly where the expectation of a common price is strongest, which is retail and ecommerce listings, and lowest where consumers already expect a quote.
What the FTC did not do
It did not ban personalised pricing. The Commission stated in the proposal that it lacks the authority to do so. It did not issue a rule, and a proposed enforcement policy statement out for public comment is not law and should not be described as one internally or to clients. It did not create obligations that bind anyone: the statement carries a disclaimer that it confers no rights and does not bind the Commission or the public.
The vote was also 2-0, which is a fact worth stating plainly rather than dressing up. It authorised the Federal Register notice, and the 30-day comment window does not begin until that notice publishes. Between now and the end of that window, nothing about the document is settled.
What to check in your own stack this quarter
None of this is urgent in a compliance sense, because there is nothing to comply with yet. It is useful as a self-test, and the work it implies is the kind that takes a quarter rather than a week if the answer turns out to be uncomfortable.
Start with the simple question of whether anything in your stack varies a listed price by individual. Personalisation engines, CDP-driven pricing rules, segment-level price experiments and anything a vendor sold as willingness-to-pay optimisation all belong in that inventory. Most teams find the answer is no, and the exercise ends there.
If the answer is yes, the next questions are the three the FTC named. Can you state that the price was personalised? Can you state the basis? Can you state the types of data used? A team that cannot answer the second and third questions internally has a data-lineage problem before it has a disclosure problem, and that usually points back at how measurement and data collection were wired in the first place.
The last piece is copy. If disclosure becomes the standard, the sentence that discloses it has to be written by someone who understands both the mechanism and the shopper, and the difference between a disclosure that reads as transparency and one that reads as an admission is entirely a content problem.
What this means for Thai marketers
Thai exporters and Shopify merchants selling to US consumers sit inside the FTC's reach, because the relevant question is where the consumer is rather than where the seller is incorporated. A Bangkok-based store shipping to buyers in the United States is selling into the market this proposed statement describes. So are Thai travel and hotel operators running willingness-to-pay segmentation against inbound US demand, which is one of the more common places where individualised pricing logic gets applied to a listing that shoppers read as a single public rate.
The more useful framing for Thai brands is as a self-test rather than a compliance deadline. Teams here are building customer data platforms and loyalty stacks under the PDPA, which already requires being able to say what personal data is collected and what it is used for. The FTC's three-part disclosure question maps onto that work almost directly: state that the price was personalised, state the basis, state the data types. A CDP that cannot answer those questions is a CDP that will struggle with a PDPA data-subject request as well.
What should not be assumed is any parallel move by a Thai regulator. The FTC announcement of 19 August 2026 concerns United States law and says nothing about Thailand, and no Thai regulatory position on personalised pricing appears in it. Treat the disclosure standard as a design principle worth adopting, not as a rule that has arrived here.
FAQ
Is personalised pricing now illegal in the US?
No. The FTC stated that it lacks authority to ban personalised pricing outright, and what it published on 19 August 2026 is a proposed enforcement policy statement open for public comment, not a rule and not law. The statement itself carries a disclaimer that it confers no rights and does not bind the Commission or the public.
Does this apply to my company if I sell from Thailand?
The announcement does not address non-US sellers, so the source does not answer this directly. What it does describe is conduct affecting US consumers, which is the basis on which the FTC operates, so a Thai merchant selling to US buyers is selling into the market the proposed statement covers. Anything more specific than that is not in the source.
Is a members-only discount personalised pricing?
Not as the FTC defined the term, provided the discount is structured as an offer rather than a price. The FTC's definition covers using personal data to set prices according to what a company believes an individual will spend. A visible list price with a named discount applied for a stated reason is a different mechanism, though this is analysis rather than something the FTC ruled on.
What about surge pricing and seasonal pricing?
The proposed statement distinguishes ordinary supply-and-demand variation from the practice it targets, along with regional and tax differences. Those are named as outside the definition, because they do not turn on what an individual consumer is believed to be willing to spend.
Do I have to change anything right now?
No. There is no obligation attached to a proposed statement that is still out for comment, and the 30-day comment period does not even begin until the Federal Register notice publishes. The useful action is an inventory of whether anything in your stack varies a listed price by individual, which is worth knowing regardless of what the FTC finalises.
The short version
The FTC drew a line on 19 August 2026 between deciding what a person sees and deciding what a person pays, and proposed that crossing it without saying so is likely a Section 5 problem. The document bans nothing, binds nobody and is open for comment for 30 days once it publishes. The part worth keeping is the three-part test, because a business that can state that a price was personalised, on what basis and from what data is a business that has its pricing logic under control whether or not the statement is ever finalised.
If you are working out where your loyalty, CDP and pricing logic actually sit on that line, that is the kind of audit the Relevant Audience team runs with ecommerce clients.
