Dynamic pricing vs personalised pricing: where the line is (and the law)
Dynamic and personalised pricing get conflated constantly. Here's the difference, the customer-perception gap, the EU/US legal landscape, and what we actually recommend operators do.
Luska Research
Pricing & Yield Team
Dynamic pricing and personalised pricing get used interchangeably in boardroom slides, and that's a problem — they're different mechanisms, with different math, different customer reactions, and different legal exposure. Here's the line, with the cases that define it.
Dynamic pricing
Same price, same moment, for everyone. The price changes over time based on supply, demand, inventory, time-to-event, and competitor moves — but at any given moment, two customers viewing the same product see the same number.
Airlines, hotels, ride-share, energy providers, and concert ticketing all use dynamic pricing. It's broadly understood, broadly accepted, and legally settled in most jurisdictions.
Personalised pricing
Different price, same moment, different customers. The price you see is a function of who you are — your past behaviour, your device, your location, the inferred willingness-to-pay from your browsing history.
It's far rarer in practice than the headlines suggest. A 2018 European Commission mystery-shopping study of 160 e-commerce sites found no evidence of consistent and systematic personalised pricing. Most "you saw a different price than your friend" stories trace back to A/B tests, geo-pricing, or device-class targeting — not true individual-level personalisation.
Dynamic pricing prices the moment. Personalised pricing prices the person. Customers and regulators react very differently to those two ideas.
Why this distinction matters
Customer reaction
Customers grumble about dynamic pricing during surge moments but accept it as fair (it applies to everyone). Personalised pricing tends to feel unfair — because the person next to you got a different number for reasons you can't see. When customers noticed different prices for the same DVDs during Amazon's 2000 price test, the company called the test a mistake, refunded the difference to 6,896 customers and said future test buyers would automatically get the lowest test price. (Amazon said the test was random, not based on who the customer was, after news reports described it otherwise.)
Legal exposure
Two big things to be aware of:
- EU Omnibus Directive (Directive (EU) 2019/2161, applied from 28 May 2022) requires sellers to tell consumers, where applicable, that a price was personalised on the basis of automated decision-making.
- Anti-discrimination law. Personalised pricing that correlates with a protected class (race, gender, age, in some jurisdictions postcode/ZIP) creates direct legal exposure even if the discrimination is unintentional. The 2015 ProPublica investigation into Princeton Review's ZIP-code-based tutoring prices, which found Asian customers nearly twice as likely to be quoted a higher price, is the textbook example.
The grey zone
What we actually recommend
For nearly every operator we work with, the answer is:
- Use dynamic pricing aggressively. It's the highest-leverage tool you have, especially on perishable inventory. See pricing strategy fundamentals.
- Use segment pricing carefully. Pricing tiers, loyalty pricing, B2B contract pricing, and student discounts are all forms of segment-level personalisation that customers accept and that work legally. Do these.
- Avoid individual-level personalisation unless you have a specific, defensible business case, explicit consent, and legal sign-off. The reputational downside is asymmetric — one screenshot on social media wipes out years of margin gain.
Where the line is moving
Three trends to watch:
- Algorithmic transparency. Regulators on both sides of the Atlantic are paying closer attention to automated pricing. Being able to explain why a price is what it is costs little to build now and is hard to retrofit later.
- Cookie deprecation is removing the raw material for cross-site personalised pricing. The remaining personalisation surface is logged-in first-party data — which is much more defensible legally but much narrower in coverage.
- Loyalty as the safe channel. A logged-in member getting a different price (you accepted T&Cs to join) is the most defensible form of personalised pricing left. Expect more retailers to migrate "personalised" pricing into loyalty programmes specifically for this reason.
Run dynamic pricing with built-in audit trails
Luska's price recommendations come with a reason and a confidence score, and wait in an approval queue until someone on your team accepts them — so every change has an explanation and a sign-off.
See AI pricing