Vendas: +86 10 6296 7490
FTC gives nine firms 45 days on personalised price files
2026-11-05博晓通

FTC gives nine firms 45 days on personalised price files

FTC gives nine firms 45 days on personalised price files imagem do artigo

As of November 5, 2026, nine companies selling flights, rides, delivery and groceries are working against a 45 day clock to explain to the US Federal Trade Commission how each price on their screens is set. The Section 6(b) orders went out on October 28 to Delta, United, JetBlue, Lyft, Uber, DoorDash, Instacart, Target and Walmart, and the commission approved them by a 2 to 0 vote. The agency says the study carries no specific enforcement purpose, but a recipient that does not answer can be taken to federal court. For an ecommerce pricing team the practical question is narrower than the legal one: could the team produce, in writing, the inputs behind a product page price.

What the orders ask companies to hand over

Known factsUnverifiedNeeds verification
Nine orders were issued on October 28, 2026 under Section 6(b) of the FTC Act to Delta, United, JetBlue, Lyft, Uber, DoorDash, Instacart, Target and WalmartThat any named recipient sets an individual price from a shopper personal dataThe pricing methods each company actually reports, and the share of sales each method covers
Each company must document the inputs, tools and procedures used to adjust or set prices and worker compensationThat loyalty program participation is treated by the agency as a personal data input rather than a published discount tierHow each company classifies loyalty pricing inside its own records and disclosures
The commission voted 2 to 0 and retains the power to compel compliance in federal court if a recipient fails to answerThat the study will end in enforcement action against one of the nineWhether the published study names individual companies or reports only aggregate findings

The wording of the release matters more than the list of names. The commission is not asking whether prices differ between shoppers, which they routinely do for reasons such as sale timing, stock and location. It is asking each company to set out the machinery behind a price, which turns a commercial question into a documentation exercise. That is the part a pricing or category team owns, not only the legal department.

Inputs, tools and procedures

Three words in the order carry most of the workload. Inputs means the data fields that entered the calculation for a given price, and whether each field varies by customer. Tools means the software or model that turned those fields into a number, including anything bought from a third party vendor. Procedures means the internal rule, approval and review that allowed the method to go live. A retailer that can name its promotion engine but cannot state which fields fed it has answered only the second of the three.

Why the 2024 study is being rerun

The commission has run this inquiry before. Its 2024 study sent orders to the consultancies and software vendors that sell pricing tools, and the release is openly critical of that choice, saying the agency learned nothing about actual pricing decisions from firms that do not set the prices. The new orders go to the businesses that do. That shift moves the exposure from vendor to operator: a supplier can describe what a tool is capable of, while only the retailer can say which data fields went into the number on a product page on a given day.

Nine recipients, four ways of pricing

The nine names are not one industry, and the order treats them differently because the object being priced is different. Read across the table and the retail rows are the only ones where the price under review is attached to a product rather than a trip or a delivery.

CompanySegmentWhat the price coversWhat the order targets
Delta, United, JetBlueAirlinesFare per seat on a routeLoyalty participation and personal data analysis; pay is salaried, so the compensation track is thin
Lyft, UberRide sharingFare per tripLoyalty status and personal data for the fare, plus shifting offers for driver pay
DoorDash, InstacartDeliveryDelivery and service fees, on top of retail item pricesFee construction, plus shopper pay built from distance and item count
Target, WalmartRetailItem price on shelf and on the product pageLoyalty pricing, personalised offers and item level promotions

The two retailers sit at the centre of the ecommerce question because their product page price is the price most shoppers can check twice, once at home and once in the aisle. A fare or a delivery fee is hard to compare across two people on the same day. A grocery item is not. That is why the retail rows are where a price monitoring routine has to be able to answer the question the regulator is asking, and where a gap between the published promotion and the price shown to one shopper would be visible first.

The evidence a product page price needs

Most pricing teams can describe their promotion calendar and their margin rules. Far fewer can reconstruct, months later, why one shopper saw a different number. The order effectively asks for that reconstruction to be possible on demand, which is a record keeping standard rather than a pricing strategy. Building it is unglamorous: the same price monitoring discipline that catches a broken promotion in a store also produces the log that shows a price came from a published rule.

Loyalty tiers are the line to draw

A loyalty price and a personalised price can look identical on a screen and are entirely different in a file. A tier price is published, available to everyone who joins, and the same for every member of that tier. A personalised price varies between two shoppers who hold the same status, on the basis of data about them. The release specifically names loyalty program participation as one factor the agency wants examined, alongside analysis of personal data by algorithms, which means a company has to be able to show which of the two produced the number. Walmart has said publicly that it prices products rather than people, and that its loyalty pricing is separate from personalised pricing; the value of that statement depends on whether its records match it.

What a disclosure has to say

The orders land next to a policy process that is already running. On August 19, 2026 the commission published a draft enforcement policy statement on undisclosed use of personal information to set a price unique to a consumer, and the comment period has been extended to November 21. The practical consequence is that disclosure wording stops being a legal footnote and becomes a record: what the page said, where it said it, and from which date. A retailer running a member price and a personalised offer on the same item needs both to be described in a way a shopper could tell apart.

What to assemble before the clock runs out

  1. List every pricing method live on the site, and name an owner for each one.
  2. For each method, record the data fields used and mark whether any field varies by individual customer.
  3. Separate loyalty tier pricing from personalised pricing in the same register, with the published terms attached.
  4. Collect the experiment record for each method, including variants that were stopped and why.
  5. Archive the exact disclosure wording, its page location and the date it went live.
FieldExample value
Pricing methodMember price on grocery, owner category manager
Data input listLoyalty tier, store catchment, stock level
Customer variation flagNo for tier price, yes for personalised offer
Experiment recordVariant, date range, result, stopped or scaled
Disclosure archiveExact wording, page location, date live

The order of the five steps is not arbitrary. A method list without a data input list produces a register that cannot answer the question. A data input list without the loyalty flag merges two legally different practices into one line. And the disclosure archive is what ties the file to what a shopper actually saw, which is the only part of the record a regulator or a customer can check without access to internal systems.

FAQ

  • Does a Section 6(b) order mean a company has broken the law?

    No. The commission describes these as wide ranging studies without a specific enforcement purpose, and the release does not allege a violation. The order is still compulsory, and non compliance can be taken to federal court, so the obligation to answer is real even when the allegation is not.

  • How is a loyalty price different from a personalised price?

    A loyalty price is published and identical for every member of a tier, so any shopper can join and get it. A personalised price differs between two shoppers on the basis of data about them, which is why the order asks companies to show which of the two produced a given number on a product page.

  • What should an ecommerce team keep for price monitoring?

    The pricing method, the data fields used, whether any field varies by customer, the experiment record with stopped variants, and the exact disclosure wording with its live date. Kept together, these show whether a price came from a published rule or from an individual assessment.

  • Why are Target and Walmart the retail names in the order?

    They are the two recipients whose business is item pricing at scale, on shelf and online. That makes the product page the place where a published promotion and an individual price can be compared directly, which is exactly the comparison a price monitoring routine already runs every day.

Sources