TL;DR. Surveillance pricing is using personal data about a shopper to set the price that shopper sees. Maryland banned it for grocery food on October 1, 2026, and the FTC proposed disclosure rules on August 19. Caveat: most documented price gaps, like Instacart’s 23% spread, were shown by testing, not proven to come from personal profiles.
Last reviewed: October 4, 2026. Reviewed quarterly.
On October 1, 2026, a Maryland law took effect that bars big grocery stores and grocery delivery apps from using your personal data to charge you more. It’s the first state law of its kind, and it landed in the same week that Walmart’s CEO published a letter saying the company prices “the product, not the person.” Both responses point at the same question: can the price you see depend on who you are?
Surveillance pricing is the practice of using personal data about a shopper, such as location, device, browsing, or purchase history, to set the price that shopper sees. In plain terms: instead of one price tag for everyone, the tag changes depending on what the seller knows about you. FindSkill.ai tracks it because it sits where AI, pricing, and privacy meet, and because both shoppers and shop owners need to understand the difference between what’s proven and what’s rumored.
Why surveillance pricing matters now
Surveillance pricing matters now because it moved from a research topic to law in about eighteen months. The Federal Trade Commission opened a study in July 2024, New York began enforcing a disclosure rule in November 2025, Maryland’s ban took effect in October 2026, and Connecticut and New Jersey have bans scheduled for 2027. Shop owners and shoppers now face real rules, not just headlines.
- Maryland, October 1, 2026: House Bill 895, the Protection From Predatory Pricing Act, bars food retailers of at least 15,000 square feet and third-party delivery apps from using personal data to set a higher price for tax-exempt food (Maryland General Assembly; Skadden, May 2026).
- FTC, August 19, 2026: the Commission voted 2-0 to seek comment on a proposed enforcement policy statement saying undisclosed personalized pricing is likely deceptive or unfair where shoppers expect a posted price not to vary (Federal Trade Commission).
- Walmart, September 25, 2026: CEO John Furner published a letter promising not to change a shopper’s price based on income, shopping history, or urgency (Walmart).
- New Jersey, July 23, 2026: the Fair Price Protection Act was signed, with its main grocery ban effective August 1, 2027 (Multistate.ai; Skadden).
Here’s the part the headlines skip. The FTC’s own August 2026 proposal says that how widely businesses use personalized pricing today “is not well understood.” The law moved faster than the evidence, which is why this guide separates what’s documented from what’s assumed.
How surveillance pricing actually works
Surveillance pricing works by feeding data about a shopper into a pricing system that estimates how much that shopper will pay, then showing a price or a promotion matched to the estimate. The inputs can be location, device, browsing, purchase history, or even mouse movements. The output can be a different price, a different “original” price, or a different deal.
The FTC’s January 2025 findings, based on documents from six pricing-technology companies (Mastercard, Accenture, PROS, Bloomreach, Revionics, and McKinsey), describe the toolbox. Those firms worked with at least 250 clients across grocery, apparel, and health and beauty. Their products could combine precise location, demographics, browsing and purchase history, device information, mouse movements, scrolling, and abandoned-cart activity, and could target prices, segment customers, or change which products appeared first.
The technical version: the pricing intermediaries sell software that sits between a retailer’s data and its price display. The retailer supplies transactions and customer attributes. The software estimates price sensitivity, often called willingness to pay, and recommends a price, discount, or product ranking. Whether the retailer then applies that recommendation to an individual shopper or to a whole store or ZIP code is the line the new laws care about.
An important point about the evidence: the FTC’s public summary used aggregated, anonymized information and hypothetical examples. It didn’t say which retailers used which data to individualize final prices. The tools exist and are sold, but nobody has published a full picture of who uses them to set one-to-one prices.
Surveillance pricing vs. dynamic pricing and its neighbors
Surveillance pricing differs from dynamic pricing mainly in who the price is tailored to. Dynamic pricing moves with demand, supply, or time and applies to everyone at that moment. Surveillance pricing is tailored to the individual’s data. The other terms in this family describe parts of the picture, which is why people mix them up.
| Term | What it means | Example | Based on your personal data? |
|---|---|---|---|
| Dynamic pricing | Price moves with demand, supply, or time for everyone | Surge fares, holiday hotel rates | No |
| Location pricing | Price depends on the store, city, or ZIP code | A Big Mac that costs more in one neighborhood | Not individual, though location is a data point |
| A/B price testing | Random shoppers see different prices so the company can learn | Instacart’s tests found by Consumer Reports | No, assignment is random |
| Personalized discount | A coupon or deal targeted from your history | A “just for you” app offer | Often yes, and legal if honest and disclosed |
| Surveillance pricing | Your data sets your price | A higher price because the system thinks you can pay it | Yes |
Maryland’s law uses the phrase “dynamic pricing” for what this guide calls surveillance pricing, defining it as a personalized price specific to a consumer based on the consumer’s personal data, including through artificial intelligence. That’s why you’ll see the same practice under several names. For a deeper look at how to tell them apart in the real world, see our consumer guide to surveillance pricing and the Maryland ban.
What the evidence shows, and what it doesn’t
The evidence shows that people have been shown different prices for identical goods at the same time, but it rarely shows why. The best controlled tests prove price dispersion. They don’t prove that personal profiles caused it. That distinction explains why companies deny wrongdoing while shoppers feel overcharged, and why lawmakers chose disclosure and consent rules.
- Instacart (Consumer Reports, Groundwork Collaborative, and More Perfect Union): 437 volunteers in four cities found about 74% of tested products at more than one price, with a largest single-item gap of 23% and a typical cart gap around 7%. Consumer Reports’ $1,200-a-year figure is a scenario applying that 7% to a household grocery estimate, not a measured loss. Instacart said the tests were random and ended them.
- Uber and Lyft (Consumer Reports, spring 2026): about 174 volunteers found every one of 30 virtual routes showed at least two price clusters, with a median gap near 50%. Both companies denied using surveillance pricing for base fares, and Consumer Reports noted it couldn’t control marketplace conditions.
- McDonald’s (Reuters, September 2026): a pricing engine recommends an “optimal price” per item per restaurant using local willingness to pay. Reuters found a Big Mac at $5.69 in one Fresno restaurant and $6.89 two miles away, but couldn’t prove the engine caused it.
The pattern is the same each time: the gap is documented, the cause is disputed. A random test, a store-level price, and a personal-data price can all look identical from the shopper’s side.
Where surveillance pricing is regulated
Surveillance pricing is regulated state by state in the United States, with no federal ban. As of October 2026, Maryland has a ban in effect, Connecticut and New Jersey have bans coming in 2027, New York requires a disclosure label, Colorado’s bill was vetoed, and California’s bill failed. The FTC’s proposal is a statement of enforcement intent, not a rule.
| Place | Status | What it does |
|---|---|---|
| Maryland (HB 895) | In effect Oct 1, 2026 | Bans using personal data to set a higher price for tax-exempt food at stores of 15,000+ square feet and delivery apps; loyalty programs anyone can join are exempt; enforced by the Attorney General, no private lawsuits |
| New York (GBL § 349-a) | In effect since Nov 10, 2025 | Requires the label “THIS PRICE WAS SET BY AN ALGORITHM USING YOUR PERSONAL DATA”; a federal court dismissed the retail industry’s challenge on Oct 8, 2025 |
| Connecticut (PA 26-130) | Signed Jun 4, 2026; operative Jul 1, 2027 | Generally bars surveillance pricing by retail sellers and food delivery services; some online sellers must disclose a personal-data price increase |
| New Jersey (Fair Price Protection Act) | Signed Jul 23, 2026; main ban Aug 1, 2027 | Bars personal-data-based grocery pricing and pauses new electronic shelf-label deployments for a year from Feb 1, 2027 |
| Colorado (HB 26-1210) | Vetoed Jun 2, 2026 | Would have restricted individualized pricing; the governor said it was too broad |
| California (AB 2564) | Failed in 2026 | Missed the final deadline; the Attorney General opened a retailer investigation in January 2026 under existing privacy law |
| Federal (FTC) | Proposed, non-binding | Aug 19 proposal: failing to disclose personalization is likely deceptive or unfair where shoppers expect fixed prices |
Dates on this topic change. Connecticut’s effective date moved three times during its 2026 session, so articles citing October 2026 or February 2027 describe earlier versions. The final operative date is July 1, 2027.
What this means for small business owners
Small business owners should treat surveillance pricing as a risk to avoid and a promise to make, not a tactic to adopt. A shop that uses a loyalty program, an ad platform, or pricing software may already be personalizing prices without calling it that, and the new laws turn that into a compliance question.
The practical split is simple. Ordinary price changes that apply to everyone are fine. A discount shoppers opt into under public rules is fine. A higher price aimed at an individual because of their data is the thing Maryland banned and other states are targeting. The Maryland law explicitly exempts loyalty, membership, and rewards programs that any consumer may voluntarily join, and prices offered to consumers who consent to share data in exchange.
The honest limit: this is a fast-moving area and the laws mostly target large food retailers today. Nothing here is legal advice. Ask any pricing or marketing vendor in writing what customer data feeds the price a shopper sees, and keep the answer on file.
The next step: the AI for Small Business course covers using AI for pricing, marketing, and customer work without crossing lines you can’t see, with the first lessons free. Our small-business spending audit for ChatGPT Finances shows the owner’s side of connecting data to AI.
What this means for e-commerce sellers
For e-commerce sellers, surveillance pricing shows up as dynamic repricing tools, personalized offers, and marketplace algorithms. The question is whether your price changes for everyone, for a segment, or for one named customer, and whether a shopper could tell the difference.
A seller using repricing software to match a competitor’s price is doing ordinary dynamic pricing. A seller showing a returning customer a higher price because the system infers they won’t shop around is doing something regulators are watching. Connecticut’s law, for example, would require some online sellers to disclose a personal-data price increase, and New York already requires a label where algorithmic pricing uses personal data.
The honest limit: marketplaces and ad platforms make many of these decisions on your behalf, and their terms change. Read what each tool does with customer data, not just what it promises to optimize.
The next step: our AI for E-commerce Operations course covers listings, inventory forecasting, and repricing with AI, and the E-Commerce with AI course gives the broader store-building foundation, with two lessons free.
What this means for marketers
For marketers, surveillance pricing is the dark side of personalization. Segmenting offers by interest is normal marketing. Using individual data to quietly raise a price is a trust and legal risk, and the line between them is disclosure and opt-in.
The workflow to protect your brand: list every place personal data touches a price or promotion in your stack, label each one as public, opt-in, or hidden, and make the hidden ones either disclosed or removed. The FTC’s proposal also warns that dressing up a personalized price increase as a “loyalty discount” may itself be deceptive, which is a copywriting problem as much as a legal one.
The honest limit: marketers rarely control pricing, but they often control the personalization layer that carries it. If you can’t answer “what data decides which offer this person sees?”, you can’t defend it later.
The next step: the Marketing Strategy course covers segmentation and offer design, and AI Ethics in Practice covers the data and transparency habits that keep personalization honest.
What this means for customer support teams
For customer support teams, surveillance pricing arrives as a complaint: “my friend paid less for the same thing.” Agents need a calm, accurate script that doesn’t promise what the company can’t verify, and they need to know which differences are legitimate.
The legitimate explanations are real and common: a member price, a store-level price, a delivery-app markup, a time-limited promotion, or a price test. A good reply acknowledges the gap, explains which of those applies if it’s known, and escalates when the customer reports a pattern. Under laws like Maryland’s, a documented pattern is exactly what a state Attorney General would want to hear about, so the escalation path matters.
The honest limit: support agents usually can’t see why a specific price was shown. Don’t guess. Capture the screenshot, the time, the item, and the device, then route it.
The next step: the AI Customer Support course covers drafting, triage, and escalation workflows with AI, with two lessons free.
What this means for freelancers and consultants
For freelancers and consultants, surveillance pricing is a reminder that quoting different clients different prices based on what you’ve learned about their budget is a form of the same thing, and it can cost you trust when clients compare notes. The fair alternative is a published rate card with clear reasons for variation.
Different prices for different scope, urgency, or volume are legitimate and explainable. Different prices for the same work because you researched a client’s funding are harder to defend. Clients talk, and a gap you can’t justify in one sentence is a gap you shouldn’t have.
The honest limit: this isn’t regulated for services the way grocery pricing is in Maryland. It’s a reputation question, not a compliance one. Write your reasons down before you quote.
The next step: the Freelancers course shows how to build AI-assisted rate cards and proposals you can explain, with the first lessons free.
Common misconceptions about surveillance pricing
Most confusion about surveillance pricing comes from treating every price difference as proof of profiling. A price gap is a symptom with several possible causes, and only some are about you. Clearing this up protects you from both overreacting and being fooled.
“Any price difference is surveillance pricing.”
No. Random A/B tests, store or ZIP-code pricing, and demand moves all produce different prices without using your profile. Consumer Reports proved different prices on Instacart, and the company said the assignments were randomized and unrelated to personal characteristics. Both statements can be true at once.
“Incognito mode or a VPN gets you the lowest price.”
There’s no broad evidence that it does. A private window doesn’t hide your IP address or login, and a VPN that changes the price mostly reveals location pricing, not individual pricing. These are reasonable parts of a comparison, not guaranteed discounts.
“A loyalty price is surveillance pricing.”
Not automatically. A member price offered under public rules that anyone can join is generally treated as a legitimate discount, and Maryland exempts it. The concern rises when your private history secretly sets a different price, or when a personalized increase is presented as a discount.
“The new laws ban all personalized pricing.”
No. Maryland’s ban is narrow: a higher price for tax-exempt grocery food, at stores of 15,000+ square feet and delivery apps. It doesn’t cover restaurants, ride-hailing, travel, or most online shopping, and there’s no private right to sue under it.
“Surveillance pricing is everywhere already.”
That is unproven. The FTC itself says current use is “not well understood,” a much more cautious line than its January 2025 release. The tools exist and are sold by pricing-technology firms, and documented price gaps exist, but the extent of true one-to-one pricing across the real economy has not been established by any published study.
Related terms
Surveillance pricing sits in a cluster of terms about AI, shopping, and data use. These neighbors are the closest in meaning, and each has its own plain-language explainer, so you can follow whichever thread matters most to your own work and your own tools.
- Agentic commerce: AI assistants that discover, compare, and buy products for shoppers.
- Amazon Rufus: Amazon’s AI shopping assistant, where recommendations and prices meet.
- ChatGPT Ads: sponsored placements for Free and Go users, with personalization controls.
- AI memory: how AI remembers you between chats, the kind of data that can feed personalization.
- Private AI: settings that stop your chats being used for training.
- Temporary Chat: a ChatGPT mode that skips history, memory, and training.
- Personal Intelligence: Gemini’s layer that reads your own Google data to personalize answers.
- AI credits: usage-based billing, another way AI changes how prices are set.
See also
This section collects the courses, glossary terms, prompt templates, and articles most closely related to surveillance pricing, grouped by type so you can jump straight to what fits your role, your experience level, and the questions you’re trying to answer.
Courses on pricing, privacy, and selling online
- AI for Small Business: automate tasks, sharpen marketing, and streamline operations
- AI for E-commerce Operations: listings, inventory forecasting, dynamic pricing, and support
- E-Commerce with AI: build an online store with AI
- Agentic Commerce for Business: AI Shopping, Decoded: what AI shopping means for sellers
- Marketing Strategy: segmentation, offers, and campaign design
- AI Customer Support: drafting, triage, and escalation workflows
- Freelancers: AI-assisted rate cards and proposals
- AI Privacy 101: settings and habits that keep your data safe
- AI Ethics in Practice: bias, privacy, and transparency frameworks
- Personal Finance: budgeting that holds up whatever the price tag says
- Claude for Small Business: set up Claude for payroll, cash, and campaign workflows
- Local AI & Privacy: run models on your own hardware
Related terms in this glossary
- Agentic commerce: AI that shops on your behalf
- Amazon Rufus: Amazon’s AI shopping assistant
- ChatGPT Ads: sponsored cards for Free and Go users
- AI memory: how AI remembers you
- Private AI: AI that doesn’t train on your chats
- Temporary Chat: a ChatGPT mode without history or memory
- Personal Intelligence: Gemini reading your own Google data
- AI credits: metered, usage-based AI billing
- Agentic calling: AI phoning businesses for you
- Answer engine optimization: being cited by AI answers
AI skills (prompt templates)
- Pricing Strategy Analyzer: pricing strategies, competitor checks, and margins
- Pricing Strategy Architect: value-based and tiered pricing design
- Pricing Power Analyzer: how far you can raise prices
- Privacy Settings Optimizer: tighten privacy across devices and accounts
- Privacy Policy Generator: GDPR and CCPA-aware privacy policies
- Data Ethics & Privacy: responsible data practices
- Grocery List Optimizer: smarter grocery lists that save money
- Return & Refund Policy Writer: clear return policies for stores
- Competitor Analysis: track competitor pricing and features
Related blog posts
- Is AI Charging You More? Surveillance Pricing, Explained: the consumer guide and a five-step price check
- ChatGPT Finances Is Free Now: A Small-Business Audit: connecting business money to AI, safely
- ChatGPT Try On: How It Works and Where Your Photo Goes: what AI shopping features collect
- The Best AI Assistant for Privacy and Trust: privacy compared across assistants
- AI Shoppers and the Etsy-Shopify Conversion Gap: how AI shoppers change seller conversion
- You Can Now Advertise Inside ChatGPT: the cost of sponsored agents for small business
- AI Pricing Compared 2026: what ChatGPT, Claude, Perplexity, and Gemini cost
Profession deep-dives
- Learn AI for Small Business: AI for owners and operators
- Learn AI for Freelancers: AI for independent professionals
- Learn AI for Entrepreneurs: AI for founders
The bottom line
Surveillance pricing is real as a technology, real as a legal category, and unproven as a widespread practice. The law is moving faster than the evidence, which means the safest position for shoppers is to check one variable at a time and keep screenshots, and the safest position for businesses is to disclose, ask for consent, and never raise a price from personal data. Treat any single price as one data point, not the price.
Frequently asked questions
What is surveillance pricing in simple terms? Surveillance pricing is when a company uses data about you, such as your location, device, browsing history, or past purchases, to decide what price to show you. The goal is to charge each person about what the company thinks that person will pay. It is different from ordinary price changes that apply to everyone at the same time.
Is surveillance pricing legal? It depends on where you live and what is being sold. Maryland has banned using personal data to set a higher price for grocery food at large stores and delivery apps since October 1, 2026. Connecticut and New Jersey have passed bans that start in 2027, and New York requires a disclosure label. Elsewhere there is no general ban.
What is the difference between surveillance pricing and dynamic pricing? Dynamic pricing changes prices over time based on demand or supply and applies to everyone at the same moment, like surge fares or holiday hotel rates. Surveillance pricing sets the price for a particular person based on that person’s data. Two shoppers seeing different prices at the same moment is the signal that something other than ordinary dynamic pricing may be at work.
Does Instacart use surveillance pricing? Consumer Reports found that about 74% of products tested on Instacart appeared at more than one price to different shoppers at the same time, by as much as 23%. Instacart called these randomized experiments, denied using personal characteristics to assign prices, and said it ended item-level price tests. The study proved different prices, not the reason for them.
Does incognito mode get you a lower price? There is no broad evidence that it does. A private window clears stored cookies but does not hide your IP address, your login, or device signals. It is a reasonable step in a price comparison, but a lower price afterward does not prove you were being targeted. A VPN that changes the price usually shows location pricing.
What should a business do about surveillance pricing laws? Separate three things: ordinary price changes that apply to everyone, discounts people opt into under public rules, and prices set from individual data. Law firms advising on the new state laws point to disclosure, consent, and avoiding personal-data-based price increases as the safer lines. Ask pricing and marketing vendors in writing what customer data feeds a shopper’s price, and talk to a lawyer.
Sources
- Federal Trade Commission, “FTC Seeks Comment on Enforcement Policy Statement Regarding Personalized Pricing,” August 19, 2026, accessed 2026-10-04. https://www.ftc.gov/news-events/news/press-releases/2026/08/ftc-seeks-comment-enforcement-policy-statement-regarding-personalized-pricing
- Maryland General Assembly, “HB 895 (2026), Protection From Predatory Pricing Act,” accessed 2026-10-04. https://mgaleg.maryland.gov/mgawebsite/Legislation/Details/hb0895?ys=2026RS
- Consumer Reports, “Instacart’s AI-Enabled Pricing Experiments May Be Inflating Your Grocery Bill,” accessed 2026-10-04. https://www.consumerreports.org/money/questionable-business-practices/instacart-ai-pricing-experiment-inflating-grocery-bills-a1142182490/
- Walmart, “A Letter From Our CEO,” September 25, 2026, accessed 2026-10-04. https://corporate.walmart.com/about/everyday-affordability/letter-from-our-ceo
- Skadden, “Maryland Becomes the First State to Restrict Surveillance Pricing in the Food Industry,” May 2026, accessed 2026-10-04. https://www.skadden.com/insights/publications/2026/05/maryland-becomes-the-first-state-to-restrict-surveillance-pricing
- Morgan Lewis, “Maryland Enacts HB 895, Becoming First State to Restrict Personalized Pricing in the Food Sector,” April 2026, accessed 2026-10-04. https://www.morganlewis.com/pubs/2026/04/maryland-enacts-hb-895-becoming-first-state-to-restrict-personalized-pricing-in-the-food-sector
- Federal Trade Commission, “FTC Surveillance Pricing Study Indicates Wide Range of Personal Data Used to Set Individualized Consumer Prices,” January 2025, accessed 2026-10-04. https://www.ftc.gov/news-events/news/press-releases/2025/01/ftc-surveillance-pricing-study-indicates-wide-range-personal-data-used-set-individualized-consumer
- Bloomberg Law, “NY Algorithmic Pricing Law Decision Provides Map for Compliance,” accessed 2026-10-04. https://news.bloomberglaw.com/ip-law/ny-algorithmic-pricing-law-decision-provides-map-for-compliance
- Multistate.ai, “State Surveillance Pricing Laws Spread as Maryland Leads the Way,” accessed 2026-10-04. https://www.multistate.ai/updates/vol-101-state-surveillance-pricing-laws
- Hunton Andrews Kurth, “Maryland Enacts First-of-its-Kind Ban on Surveillance Pricing for Grocery Sales,” May 2026, accessed 2026-10-04. https://www.hunton.com/privacy-and-cybersecurity-law-blog/maryland-enacts-first-of-its-kind-ban-on-surveillance-pricing-for-grocery-sales
- Google Ads search data and keyword difficulty via DataForSEO (US), accessed 2026-10-04. https://dataforseo.com/
