Photo By: Vitaly Gariev
For years, consumers have grown accustomed to algorithms deciding what they see.
Netflix recommends what to watch. Retailers personalize product suggestions. Loyalty programs deliver different offers to different shoppers. Advertising platforms decide which promotion appears on which screen.
But there is a potentially more consequential frontier for personalization: deciding what someone should pay.
As artificial intelligence gives companies increasingly sophisticated ways to analyze purchasing behavior, businesses have more information with which to estimate willingness to pay. That capability is forcing some of America’s largest companies to establish boundaries around how pricing technology should be used.
Walmart CEO John Furner recently made his company’s position unusually explicit. Despite Walmart’s expansion of digital shelf labels and AI across its stores, Furner said the retailer would not use customer identity to determine prices, summarizing the philosophy as: “We price the product, not the person.”
The distinction may become increasingly important as companies discover just how precisely AI can understand their customers.
Personalized Commerce Has a Pricing Boundary
Consumers already participate in an enormous amount of personalized commerce.
Two people visiting the same website may receive different product recommendations. Loyalty members can receive targeted coupons. Advertisements are selected using information about audiences and their behavior.
Personalized pricing changes the equation.
There is a meaningful psychological difference between an algorithm concluding, “This person is likely to want this product,” and concluding, “This person is likely to pay more for this product.”
Regulators are beginning to examine that boundary.
The Federal Trade Commission recently sought public comment on a proposed enforcement policy concerning personalized pricing, which it defines as using personal data to set prices according to what a business believes an individual consumer is willing to spend.
The agency’s proposal doesn’t seek to prohibit personalized pricing in every circumstance. Instead, it focuses heavily on transparency, warning that businesses could run into existing consumer-protection laws when customers reasonably expect prices to be uniform but aren’t told personal data is influencing what they pay.
That turns personalized pricing into more than an optimization problem. It becomes a trust problem.
Not Every Changing Price Is Personalized
The distinction matters because companies have adjusted prices according to market conditions for decades.
Airfares fluctuate. Hotel rooms cost more during periods of high demand. Retailers mark down excess inventory. Restaurants can charge different prices across markets.
That’s dynamic pricing. A company might also divide customers or markets into broader segments and use different pricing or promotional strategies across them.
Personalized pricing goes further by incorporating personal data into determining what a particular consumer may be willing to spend.
AI makes that increasingly feasible because companies possess enormous amounts of behavioral information.
Purchase history can reveal brand loyalty. Promotion history can reveal price sensitivity. Browsing behavior can indicate purchase intent. Abandoned carts, product preferences and transaction frequency provide additional signals.
Combined, those signals could allow increasingly precise estimates of willingness to pay.
The technological question is becoming: How much can a company know?
The strategic question may be more difficult: How much should it use?
The Highest Possible Price May Carry a Hidden Cost
The intuitive business case for personalized pricing is straightforward. If one customer will pay $50 and another will only pay $40, why charge both $40?
But maximizing the value of an individual transaction isn’t necessarily the same as maximizing the value of the customer relationship.
A higher price can affect conversion, purchase frequency, retention and future price sensitivity. If customers discover they are paying more based on information collected about them, the effect could extend to perceptions of the brand itself.
Kapnova, an agentic revenue and profit optimization system for consumer brands, approaches pricing as one part of a broader commercial decision rather than an isolated transaction. Its systems can evaluate decisions against outcomes including revenue, gross profit and contribution margin.
That distinction matters as pricing becomes more personalized.
An algorithm may determine that a particular customer is willing to tolerate a higher price. A broader decision system still has to determine whether charging it creates the best economic outcome. Sometimes those answers may differ.
Transparency Could Become Part of the Pricing Decision
The FTC’s proposed approach introduces another variable companies may have to consider: whether they are comfortable explaining the strategy.
Where consumers reasonably expect prices not to vary based on personal information, the agency’s proposed policy says businesses engaging in personalized pricing should clearly disclose that prices are personalized, the basis for the personalization and the types of data involved.
That creates a useful test for companies considering how far to take AI pricing.
Would the strategy still make business sense if the customer understood how the price was determined?
For some uses of personalization, the answer may be yes. A company could potentially use sophisticated pricing analytics to offer lower prices to highly price-sensitive consumers, improve accessibility or design targeted promotions.
Research from the University of Chicago Booth has previously demonstrated that personalized pricing doesn’t automatically mean every consumer pays more. In one field experiment using machine learning, more than 60% of consumers benefited from lower personalized prices than they would have received under an optimized uniform price.
That complexity is precisely why the debate isn’t as simple as personalized pricing being inherently good or bad.
AI Is Making Pricing a Choice, Not Just a Calculation
Companies will become increasingly capable of estimating what different consumers are willing to pay. That doesn’t mean every company will use everything it knows.
Walmart’s decision demonstrates that businesses can establish their own boundaries even when technology might permit them to go further. Regulatory scrutiny could establish additional ones.
The next era of pricing may therefore require companies to optimize across several variables simultaneously: revenue, profit, consumer expectations, transparency and trust.
AI can help answer how much someone might pay.
The more consequential decision is whether a company believes it should charge them that amount.