Five shoppers opened Instacart in the same hour, at the same store, and added the same dozen eggs to their carts. They saw five prices: $3.99, $4.28, $4.59, $4.69, $4.79. The market did not set any of those numbers. An algorithm did, using each shopper's data to guess the most they would pay.
"The market sets the price." Most operators over 45 have heard that line so many times it sounds like a law of physics. It is not a law. It is not even true. The number on your screen is now a personal estimate, built from your browsing, your location, and your buying patterns, without your knowledge.
The Research Is Clear on This
In September 2025, Consumer Reports and the Groundwork Collaborative ran a measured test. They sent 437 shoppers across four cities to add the same items to their Instacart carts and record the prices they saw. Close to three out of four grocery items were shown at more than one price. The gap between the high and the low for the same product ran as wide as 23 percent. For a family of four, those gaps added up to $1,200 a year in added cost.
Same store. Same product. Same hour. Five price tags on one carton of eggs.
The old line does not fail people. It fails the system they are trying to run. If you set your own prices based on what you think "the market" charges, and the market is now five numbers for the same item, your frame is broken at the root.
The Flaw Is Not Greed. The Flaw Is Asymmetry
Grace Gedye, a senior policy analyst at Consumer Reports, named the force at work. The seller knows your data. You do not know anyone else's price. That gap is not a side effect. It is the business model.
Retailers track how often you search for an item, whether your cursor hovers on the buy button, where you live, what your income looks like, and how fast you check out. They feed all of it into one output: the most you will pay.
The problem is not that companies want more money. The problem is that data-driven pricing replaced market pricing, and nobody told the buyer. The FTC's own report in early 2025 warned this goes past eggs. A new parent looking for baby formula. A buyer in a food desert with no other store. A shopper who never compares. The algorithm finds the people least likely to push back and prices into that gap. FTC Chairman Andrew Ferguson said it flat: when you see a listed price, you expect it to be the same for everyone. It is not. It is the retailer's best guess at what you, alone, will accept.
Most people treat this as a personality flaw, as if they should have been more careful. It is a system flaw. The asymmetry is built into the model by design.
The Better Principle
Price is no longer found by the market. It is assigned by whoever holds the data. Once that is clear, three moves follow from it.
Move 1: Run the two-screen test
Pull up the same product on your phone and on a laptop in a private browser with cookies cleared. Use a different address if the tool allows it. Compare the two numbers. This takes less than two minutes, and what you see will tell you more about how your vendors price than any sales call ever did.
That small test asks something of you: the will to look at a number you were not meant to see.
Move 2: Price your own work on structure, not on signal
If the price your customer sees can shift based on data they never chose to share, then "market rate" is no longer a fixed point you can anchor to. Set your prices on your costs, your margin target, and your delivery time. Not on what a search result says the going rate is. The going rate is now a range of one, built for the person who searched.
Move 3: Watch the ground shifting beneath your pricing page
Instacart shut down its AI pricing tests after the study went public. Companies do not kill tools that make money unless the tools did what the study said. Beyond that, more than 40 bills have been filed across 24 states to rein in this kind of pricing. Maryland now charges up to $10,000 per first-offense violation and up to $25,000 for repeat violations. In mid-2026, the FTC proposed an enforcement policy that would treat this practice as a trigger for federal scrutiny under Section 5 of the FTC Act. If you sell anything online, the legal ground beneath your pricing is moving. Know where it sits before it moves again.
What the System Shows
Running these three checks over the next few weeks does something the old advice never did:
You see the real spread between what you pay and what the next buyer pays for the same thing.
You find out which of your own vendors are running data-driven pricing on you.
You learn whether your own pricing holds up when a customer runs the same two-screen test on your site.
You stop treating "market price" as a fact and start treating it as a claim that needs proof.
Three Questions Worth Asking
At the end of a month, ask three things.
→ Which of my costs shifted without a clear cause, and was the shift tied to my own data trail?
→ Which price did I accept without checking whether someone else saw a different number?
→ Where is my own pricing built on a "market rate" that no longer exists as a single figure?
That is the gap between advice that sounds right and a system that proves itself.
Where You Stand
The old line said the market sets the price. The measured data show the algorithm sets it, one buyer at a time, using information the buyer never agreed to hand over. You sit on both sides of this. You are the buyer being profiled. You are the seller whose pricing sits on ground that is shifting fast.
The price on your screen is not the market. It is a mirror of what the system believes about you.
