If you’ve ever compared prices with a friend and noticed a discrepancy — same product, same site, different prices on different devices — you may have encountered surveillance pricing without knowing it.
Surveillance pricing is the practice of using personal data to set individualized prices in real time. Your browsing history, location, device type, purchase history, income signals, and behavioral patterns all feed into systems that determine not just what you’re offered, but exactly what you’re charged. The price isn’t a market rate. It’s a price calculated specifically for you.
This isn’t a niche academic concern. US senators held a formal hearing on the practice in August 2026. More than 40 bills have been introduced across US states this year alone to regulate or ban it. The FTC has issued a major report on it. Retailers are experimenting with dynamic electronic price tags in physical stores. The practice is mainstream — most people just haven’t had a name for it until recently.
How Surveillance Pricing Actually Works
The mechanism varies by industry, but the core logic is consistent.
A company collects data points about you — some directly (account registration, purchase history, in-session browsing), some from third-party data brokers (income proxies, neighborhood demographics, credit scoring), and some from behavioral inference (how long you linger on a product page, how many times you’ve visited the site, whether you’ve abandoned a cart).
This data feeds into a pricing algorithm — increasingly an AI model — that assigns you a willingness-to-pay score. The system then shows you a price optimized for extraction: high enough to maximize revenue from you specifically, but low enough that you’re likely to complete the purchase.
The practice is most entrenched in sectors where dynamic pricing has been normalized for years: airline tickets, hotel rooms, ride-sharing. It’s now expanding into retail, insurance, subscription services, and consumer software.
What Data Feeds These Systems
Understanding surveillance pricing means understanding the inputs. They’re more varied than most people expect.
Device and access signals. The type of device you’re using is one of the oldest signals. Apple device users have historically been shown higher prices at some travel sites, based on the inference that iOS users have higher average incomes. Browser, operating system, and even screen resolution can all contribute to the pricing fingerprint.
Location data. Your approximate location — derived from IP address or GPS — is used to infer local purchasing power. Users in high-income ZIP codes may see different price points than users in lower-income areas, even for digital goods with no geographic cost component.
Behavioral session data. How you interact with the site is tracked. How many times you’ve visited. Whether you’ve added something to a cart and abandoned it. Whether you’re arriving from a competitor’s page. A user who has repeatedly returned to view an item signals high intent — and may be shown a lower price to close the sale, or a higher price if the algorithm decides they’ll pay regardless.
Third-party data broker inputs. Many pricing systems augment their own data with purchased signals: income range estimates, credit score proxies, life stage indicators (recent home purchase, recent divorce, new child), and aggregated behavioral profiles assembled by data brokers. You never consented to this data flowing here. It got there through intermediaries, in ways that are technically legal in most jurisdictions.
Account history. Long-tenured subscribers may see renewal pricing that differs from new signup offers. Users who’ve previously cancelled and returned are flagged. In some cases this produces genuinely good deals; in others it’s a mechanism for charging loyal customers more, banking on their friction with switching.
Insurance Is Where This Gets Consequential
Retail surveillance pricing is annoying. Insurance surveillance pricing affects how much you pay for coverage that protects your health, home, or car.
Auto insurance telematics programs — the kind where you install an app or device to “earn discounts” — track driving patterns, time of day, hard braking, and speed in fine-grained detail. The data generates individualized premium rates. If you drive late at night or brake hard in urban traffic, your rate climbs — not because you’ve made a claim, but because a behavioral profile infers you’re higher risk.
Life and health insurance pricing is incorporating wearable data, purchase histories, and credit scoring in jurisdictions where this is permitted. Your premium isn’t a commodity rate. It’s a function of the data that’s accumulated about you from sources you may not be aware of.
The asymmetry here matters: companies hold rich profiles. You hold almost none of that information about how you’re being priced. There’s no mechanism to see your willingness-to-pay score, request a correction, or opt out of the model that set your rate.
The App Subscription Problem
Subscription apps are a growing surveillance pricing vector that gets less attention than retail.
Many consumer apps — streaming services, productivity tools, VPNs, fitness apps — now use A/B testing and behavioral segmentation not just to optimize conversion flows but to set baseline prices. A new user downloading an app may see a substantially different price than an existing user who’s been tracked across devices for months.
In-app purchase pricing pioneered this in mobile gaming. Companies became expert at identifying high-spending users — called “whales” in industry parlance — and targeting them with more aggressive pricing and upsell flows. The same logic has spread to subscription apps across categories. If the algorithm has inferred you’re a high-value target, you may see different prices than someone the model classifies as price-sensitive.
This is difficult to detect because prices are personalized and non-public. There’s no posted menu to compare against. A/B testing looks identical to dynamic pricing from the user’s perspective. Companies don’t disclose the signals that go into pricing models.
Why Regulatory Progress Is Slow
Surveillance pricing occupies an uncomfortable legal space. Price discrimination has long been regulated under antitrust law, but algorithmic personalization is treated differently — partly because the legal categories predate behavioral AI, and partly because the practice generates significant revenue for large, politically influential companies.
Regulatory disclosure requirements are still nascent. New York’s Algorithmic Pricing Disclosure Act was among the first enacted, requiring certain disclosures when algorithmic pricing is in use. Maryland followed with the Protection from Predatory Pricing Act. California, Illinois, Vermont, and Connecticut all have active bills at various stages. Federal legislation is proposed but not enacted.
The practical effect for consumers is that the practice is almost certainly happening to them, in ways they can’t see, with no immediate remedy available.
What You Can Actually Do
You can’t opt out of surveillance pricing entirely. The inputs are too distributed, the practice too embedded in standard commercial infrastructure. But several choices reduce your exposure meaningfully.
Use private browsing for price research. Shopping in a private window prevents the site from reading prior visit counts and limits session-specific signals. It doesn’t address account history or third-party data, but it removes behavioral tracking within the current session.
Compare prices across devices and states. Check prices while logged out versus logged in. Try a browser you rarely use, or a different network. If prices differ meaningfully, you’ve found direct evidence of personalized pricing. Screenshots make useful evidence if you want to escalate.
Be cautious about behavioral data flows. The apps you use daily contribute to the data broker ecosystem that feeds pricing models. Apps with location access, contact sync, or integration with financial accounts generate signals that may end up in pricing algorithms you’ll never interact with directly. Reducing which apps collect which data reduces your pricing profile — slowly, but persistently.
Exercise data broker opt-out rights. California’s Delete Act, Texas’s TDPSA, and equivalent state laws in around 20 US states allow residents to request data deletion and opt out of data broker sales. Exercising these rights doesn’t remove you from all pricing systems, but it reduces the richness of the profiles available to purchase. Several services (Privacy Bee, DeleteMe) aggregate these opt-out requests for a fee.
Choose services with fixed, published pricing. Transparent pricing is a feature worth considering as a selection criterion. Services that publish identical rates to all customers and don’t run behavioral targeting infrastructure are making a structurally different product choice. You may pay more than someone who got a personalized discount, or you may pay less — but you know what you’re paying before you commit.
The Root Cause
Surveillance pricing is an effect. The cause is behavioral surveillance at scale — the ambient collection of behavioral signals that funds the advertising industry, trains AI models, and is now being applied to price optimization.
Every app that runs on an advertising or data monetization model is a node in this infrastructure. The data doesn’t stay with the app that collected it. It flows to brokers, to ad networks, to partners, and eventually — through chains of vendors and aggregators — into pricing systems you’ll never see and never agreed to participate in.
The structural alternative is simpler to describe than to implement: services that don’t collect behavioral data for monetization have nothing to contribute to surveillance pricing pipelines. Fixed pricing, no advertising, no data sales. The price you’re offered is the price everyone pays. There’s no scoring model, no willingness-to-pay estimate, no segment you’re silently placed in.
That model doesn’t solve surveillance pricing as an industry-wide problem. It is, however, a coherent individual choice — and the more clearly you understand what the alternative model involves, the easier it is to make that choice deliberately.
Knowing You’re Being Profiled
The most important shift is simply recognizing that “the price” is a fiction in many contexts. When you shop online, use an app, or renew a subscription, you’re not necessarily seeing the market rate. You’re seeing the rate an algorithm has calculated for you based on a profile you’ve never reviewed.
Once you see that, the question of which services to use, and which to avoid, looks slightly different. Not every personalized price is unfair. But you should know when you’re in a personalized pricing environment, and you currently have almost no way to tell.
That information asymmetry is what the pending regulatory changes are trying to close. Until they do, the only practical protection is knowing how the system works.