Algorithmic Pricing Terminology Research Note

Algorithmic Pricing Terminology Research Note

Question

Which pricing terms describe different mechanisms, and what evidence is required before Venture Step uses each term?

Working distinctions

The categories can overlap. A company could run an A/B test of personalized discounts. A location-based market price can be dynamic without using an identified person's history. Editorial language should name only the mechanism supported by evidence.

FTC usage

The FTC's January 2025 staff study says pricing intermediaries can use precise location, browsing, shopping history, demographics, mouse behavior, and other consumer information to tailor prices and offers.

The study reviewed documents obtained through 6(b) orders from eight intermediaries. It establishes market capability and examples in aggregated form. It does not prove that every retailer or pricing experiment uses every described input.

Legal language

The New York attorney general's January 2026 announcement says the state's disclosure act applies when personal data is used to set individualized prices. The statutory wording and enforcement status need current legal review before publication.

Do not assume that the FTC's policy term and every statute use identical definitions. The public explainer provides conceptual education, not legal advice.

Editorial rule

Observed simultaneous differences establish variation. They do not establish personal-data causation. A company statement that assignment was randomized is relevant counterevidence but not an independent audit.

Prefer a precise factual sentence over a stronger label when the mechanism is uncertain. “Coordinated shoppers saw different prices for the same item” is a supported observation. “The platform used each person's willingness to pay” requires evidence about the inputs and rule.