Two people can visit the same website, look for the same thing, and end up with different prices, different discounts, or even a different view of what is worth buying.
But that does not mean every price difference is personalized.
The useful way to understand online pricing is to separate three things that can change: the market, the shopper, and the offer. Sometimes the price changes because circumstances changed. Sometimes information about the shopper can influence the result. And sometimes the listed price stays exactly the same while the deal or shopping experience changes around it.
That distinction matters because a changing price is easy to notice, but the reason behind it usually is not.
Sometimes the market changed, not the price for you
Suppose you check a hotel room, airline ticket, delivery service, or product in the morning and see $80. You return later and see $95.
The only thing you know for certain is that the price changed.
A seller may adjust prices because demand increased, inventory fell, a promotion ended, competitors changed their prices, or local market conditions changed. Pricing systems can make these adjustments automatically, sometimes very frequently.
This is generally called dynamic pricing.
The important point is that dynamic pricing does not require the seller to know anything special about you. Everyone shopping under the same conditions might receive the new price.
The Federal Trade Commission's proposed August 2026 policy on personalized pricing makes this distinction explicitly. It separates pricing based on personal information from ordinary differences caused by factors such as supply and demand, taxes, local market conditions, or characteristics of the transaction.
So seeing a price change after you return to a website is not evidence, by itself, that the website recognized you and decided you would pay more.
But sometimes the shopper can become part of the calculation
Personalized pricing is different.
Instead of asking only:
What should this product cost right now?
a system can also ask:
What price should be presented to this particular consumer?
In August 2026, the FTC proposed defining personalized pricing around the use of personal data to set an individual's price based on estimates such as how much that person may be willing to spend. The proposal is not yet final, but it reflects a narrower distinction between ordinary price changes and prices based on information about the consumer.
Earlier FTC research found commercial systems capable of using information such as browsing and transaction history, location, purchases, demographics, website activity, and inferred preferences. Some tools could combine those signals with market conditions to generate targeted prices.
That means personalized pricing is not merely a theoretical possibility. Commercial systems capable of doing it exist, and peer-reviewed field research has independently demonstrated that machine-learning systems can construct and test individualized prices in real commercial settings.
But this is where an important limit enters the picture.
The evidence that personalized-pricing technology exists is much stronger than the evidence showing how often consumers actually encounter it.
As of August 2026, the FTC itself said the extent to which businesses currently use personalized pricing is not well understood.
So it would be justified to say that businesses can personalize prices. It would not be justified to say that most online retailers routinely calculate a different price for every shopper.
The price may not be the part that changes
Focusing only on the number beside the product can also miss something important.
Imagine two shoppers looking at the same $50 item.
One receives a 20 percent coupon.
The other receives no coupon.
The posted price is identical, but the actual offer is not.
FTC staff found systems capable of using browsing and transaction information to determine who receives promotions or discounts. One example described a retailer potentially withholding a promotion from regular customers expected to buy anyway while offering it to less frequent customers who might otherwise leave.
In that situation, personalization does not mean charging one person a higher list price. It means deciding who receives an incentive and who does not.
The same principle applies to loyalty offers, retention discounts, and other targeted promotions.
So "same price" does not necessarily mean "same deal."
Even the store you experience can be different
The underlying prices can remain unchanged and the shopping experience can still be personalized.
Online stores decide which products to place first in search results and recommendations. FTC staff found tools capable of using customer segments or real-time behavior to influence which products receive prominence.
Imagine two people searching for a baby thermometer.
Both may have access to exactly the same catalog. But one person could be shown lower-priced products first, while another is presented with premium products more prominently.
Nothing about the price of an individual thermometer has changed.
What changed is which prices the shopper is steered toward first.
That broadens the original question. Online personalization is not only about whether two people see different numbers beside the same product. It can also affect the discount they receive or the products they are encouraged to consider.
Location can look like personalization when it is not
Location makes the picture more complicated.
Two shoppers in different places may see different prices because stores face different competitors, inventory levels, taxes, delivery costs, regulations, or regional pricing strategies.
FTC research describes a system capable of using nearby competitors' prices to calculate different store-level prices and then showing online visitors the price associated with their nearest inferred store.
That is not necessarily personalized pricing in the sense of estimating what an individual is willing to pay.
But location can also be part of a customer's profile.
The difference depends on how the information is being used, not simply on whether two people in different places see different prices.
What can you actually conclude when a price changes?
Very little from the price change alone.
If a product rises from $80 to $95, possible explanations include changing demand, inventory, timing, geography, promotions, seller differences, account status, experimentation, or personalization.
Even two people comparing prices at roughly the same moment may differ in ways they cannot see.
That is why claims such as "the price went up because I searched twice" require more evidence than the price change itself. Browsing behavior can be an input into some commercial systems, but that does not establish that repeated searching caused a particular increase.
The same is true of assumptions about incognito mode. If a price difference is driven by inventory, local markets, account information, or other signals, changing browser mode does not reveal or control the underlying mechanism.
The mental model to keep
Online pricing is not one system with one explanation.
A useful way to think about it is this:
The market can change the price.
Information about the shopper can sometimes change the price or offer.
And the shopping experience itself can be personalized even when the price does not change.
Those mechanisms can also operate together.
Commercial personalized-pricing systems exist. Research has demonstrated that individualized prices can be produced. But the current evidence does not establish that individualized pricing is routine across online shopping, and the FTC says its present prevalence is still not well understood.
So when two people see different outcomes online, the difference is real.
What usually remains hidden is why.
The price can depend on the product, the moment, and sometimes the shopper. And even when the price is the same, the offer may not be.
