Consider a simple hypothetical: two renters book the same car class at the same airport location for the same dates, and one is quoted $6 a day more. If she asks why, can anyone at the counter or at headquarters show her how that rate was set?

Rates adjusted for fleet availability, utilization, local events, and seasonal demand are ordinary rental pricing, and the proposed, pending, or passed personalized-pricing rules don't treat that kind of demand-based pricing as the problem. Legislators and regulators are concerned when a rate is set, even in part, from personal data about the person booking.

To explain a personalized rate, an operator should retain the data used, the rule or model applied, and the base rate before personalization. Those records provide a practical starting point for answering customer questions and assessing the pricing practice against applicable requirements.

This article offers operators practical guidance as federal and state requirements develop.

What Regulators Are Asking For

On Aug. 19, the Federal Trade Commission proposed an enforcement policy statement on personalized pricing. Where customers reasonably expect one price for everyone, a business that uses personal data to set an individual price would be expected to disclose clearly that the price is personalized, the basis for it, and the types of data used.

The statement is not final. Among the inferences the FTC lists as raising concern when used to raise an undisclosed price is a consumer traveling for a funeral or other urgent obligation. In a hypothetical rental example, a booking flow could read a last-minute, one-way search from an airport as that kind of signal.

Since Nov. 10, 2025, New York state law has required a business that sets a price algorithmically from a New York consumer's personal data to display the words "THIS PRICE WAS SET BY AN ALGORITHM USING YOUR PERSONAL DATA."

That requirement can apply to a personalized loyalty discount as well as to a higher price. In June, the legislature passed the One Fair Price Act, which would go further than disclosure and ban pricing based on personal data, with exceptions for loyalty and membership discounts that meet the bill's conditions. The bill awaits gubernatorial action.

Connecticut's law, which takes effect July 1, 2027, will restrict certain personalized pricing, apply to prices set for groups of consumers as well as individuals, and exclude certain loyalty programs. Both draw the line an operator will have to work with: a loyalty or membership discount on stated terms is treated very differently from a price built from a renter's profile. These rules differ in what they require, and an operator in several states will need counsel's view on each.

Operators serving several states can use a common process to document how they calculate personalized rates, then check each pricing practice and disclosure against the relevant state's requirements.

Where Personal Data Enters a Rate

Personal data can determine which discounts a renter qualifies for, which additional products are offered, and the prices quoted for them. Examples include loyalty discounts, corporate rates, offers to previous customers and prices for upgrades or protection products.

A published member rate or corporate discount gives renters a stated basis for the price. Questions become harder to answer when a price reflects the phone used to book, whether the renter booked through an app or website, repeated searches, or demographic information such as household income. If those details affect the quote, the operator should be able to explain how they affected it and why they were relevant.

Explaining a Discount Versus Rebuilding a Rate

Deterministic segmentation groups renters using explicit eligibility rules and applies a defined rate adjustment or offer to each group. The base rate can still change with demand; the same eligibility information and rule version produce the same discount decision.

This gives the operator a specific explanation for the discount: which criteria the renter met and which rule determined the reduction. Data and models can still help identify groups that respond differently to price, while the approved rule determines the production discount.

For example, an operator could offer loyalty members a 10% discount on the base daily rate after three direct rentals in the previous 12 months. Eligibility would be checked against its rental and membership records, with the discount rule recorded as version 2, effective October 1. The customer-facing explanation could read: “10% loyalty discount for completing at least three direct rentals in the past year.” This record explains the discount, but reproducing the full quote also requires the base daily rate that applied at the time.

To rebuild what the renter was quoted a year later, the operator also needs to keep, for each booking: the base rate at the time of the quote and the version of the system that produced it; the inputs that placed the renter in the segment, as they stood at booking; the rule version applied; any other adjustments, such as a corporate code, fees or taxes; and the final rate and the wording the renter saw.

With that, the rate can be recalculated rather than approximately explained.

Compare that with a model that combines device type, home ZIP code, a late-night search, and repeated look-ups for the same dates, decides the renter isn't price-sensitive, and adds 8%. A model can be reproducible too, if the operator logs every input as it stood at booking, keeps the exact model version that ran, and can rerun it.

Without those records, there's no evidence why this renter paid 8% more. Even with them, a rerun shows the output, not a criterion the renter could check. Explainability tools can estimate afterward which variables mattered, but different tools can produce different accounts of the same rate, making it hard to show that the explanation actually set the price.

Which Data Can Move a Rate

An operator's own rental and loyalty history is the easiest data to explain, because the renter generated it and can check it. Demographic data is harder to defend: in a January 2025 Consumer Reports survey, 76 percent of Americans opposed loyalty discounts based partly on characteristics such as age and income.

Connected-vehicle data carries particular risk, and access to it was one of three priorities ACRA members took to Capitol Hill in September. The FTC's final order against General Motors and OnStar in January shows how the agency views that data: GM must obtain affirmative express consent before collecting, using, or sharing connected-vehicle data, and is barred for five years from sharing driving and location data with consumer reporting agencies. Letting driving behavior move a rental rate would invite the same questions.

Operators should approve third-party data, such as broker data or signals supplied by a booking vendor, before it can move a rate, with a record of its source and what the renter agreed to.

Where to Start

The first question a regulator is likely to ask is simple: who pays more than the base rate, who pays less, and why? Few operators simply mark rates up; more often, discounts for some renters sit alongside higher rates for others.

An operator should know which outweighs the other before anyone asks, and should prepare that analysis with counsel, because internal documents on who pays more carry weight in an investigation. None of this requires switching off the current revenue management system:

1. Map who pays more and who pays less. Compare each segment's payment to the base rate, how much, and on what criterion.
2. Identify each decision that can change a rate or offer. Record the data behind it and the team that owns it, including
3. marketing and e-commerce. For each input, note whether it is your own data or a third party’s, a recorded fact or a model’s inference, and whether it can raise a rate or only lower it.
4. Write each personalized discount down as a rule. State the criteria, the data source, the treatment against a defined base rate, and an effective date. Revenue management should own the rule; legal and privacy should approve its data. Any input that is inferred, comes from a third party, or can raise a rate needs explicit sign-off.
5. Run the rules alongside your current system. Compare rule-based rates with your revenue management system’s, segment by segment, against an acceptable revenue difference set in advance. Generate the renter-facing disclosure from the same record that set the rate.

Whether a practice is lawful in a given state remains a question for counsel. What an operator can check is whether it can answer who pays more, who pays less, and why, for every segment, and rebuild one past rental from the record alone, to the cent: the base rate, segment, rule version, adjustments, final rate, and the sentence the renter saw.