Price Does Not Decide the Amazon Buy Box
We turned Buy Box Checker on in late January 2026 and have not turned it off since. What follows is what the data says after seven months of checking the same products from many places at once.
It started with one brand that had no competitors
Buy Box Checker did not begin as a competitive intelligence tool. It began with a fertilizer brand that sold its own product, on its own listing, with nobody else on it. There was no hijacker to catch and no rival to outbid. The question was simpler and more unsettling: were they actually purchasable in Florida?
Their Florida listing kept getting dinged. Something in that state was suppressing the offer, and Seller Central was not showing it, because Seller Central reports the Buy Box as one blended national number. A national average cannot answer a question about one state. So we built something that checks the same product from many delivery addresses and reports each one separately.
That origin still shapes what the tool is for. Plenty of sellers use it to watch competitors. But the first job was knowing whether you can be bought at all, in a specific place, right now. Amazon does not surface that.
What we have checked so far
The sample has grown considerably since that first brand. It now spans four marketplaces, and it is checked continuously rather than sampled.
831,091
Buy Box checks logged
since late January 2026, across the US, UK, Canada and Ireland
123,453
competitor offers captured
recorded when a check found another seller on the listing
Those figures are live. They are read from the database each time this page regenerates. That is also why the video above quotes numbers that no longer match exactly. It was right when it was recorded. The sample simply kept growing while it was being edited.
One boundary worth stating before the findings, because it is the kind of thing that quietly makes a statistic mean something other than what it says. The two totals above count every marketplace we monitor. The five findings below are measured on amazon.com only. We track the UK, Canada and Ireland too, and those checks are deliberately held out of these particular aggregates rather than blended into a single number that would describe no marketplace in particular.
Finding one: location changes the answer
The first thing that becomes obvious when you check a product from several cities on the same day is that you frequently get different answers. Not different prices. Different Buy Box status. Owned in one city, lost or suppressed in another, at the same moment.
18.1%
of products checked in two or more locations split the Buy Box
measured on the most recent complete day. Today is excluded until its scrape finishes.
About a fifth, day after day. That is not a rare edge case, and it is invisible to any tool that reports one national percentage. If you are looking at a healthy account-level Buy Box number, it can be hiding a city where you have not held the Buy Box in weeks.
Finding two: price does not explain it
This is the result that surprised us most, and it is the reason this post is titled the way it is. The intuition every seller has is that the cheapest offer wins. So we took the products whose Buy Box status differs from city to city and asked a narrow question: are they winning in the city where their price is lowest?
Mostly, no. Only a minority of them are winning where they are cheapest. A larger group is winning at a higher price than the city where they are losing. And a substantial group shows no price difference between the two at all, which means price cannot be the variable that separated them.
15.4%
of split products win where they are cheapest
Of 52 products whose Buy Box status differs by city, 8 win where their price is lowest, 26 win at a higher price than where they lose, and 18 show no price difference at all.
The sample is small and we are not going to pretend otherwise. A few dozen products is an indication, not a law, and we say so on the research page where the figure lives and updates. But it is enough to retire the idea that price alone decides this. The likeliest explanation for the rest is delivery speed: inventory sitting closer to that shopper arrives sooner, and a faster promise beats a slightly cheaper one.
If you want the mechanism rather than the measurement, we wrote about how the Buy Box algorithm actually picks a winner, including why we refuse to publish the precise factor weights you will find elsewhere.
Finding three: the gap between cities is wide
Win rates are not evenly spread across the country. When we rank the cities in our sample by the share of tracked product-locations currently winning the Buy Box, the best and worst are separated by a margin far too large to be noise. Same catalogue, same moment, different city.
16.3 pts
win-rate spread between Los Angeles and Denver
Cities backed by fewer than five distinct brands are excluded, because a rate built from one brand describes that seller rather than the market.
We are careful about which cities appear here. A city backed by a single brand would publish that one seller's performance rather than an aggregate, which would be neither useful nor fair to them, so it is excluded.
Finding four: suppression is as common as losing
Sellers watch for a competitor taking the Buy Box. Far fewer watch for it disappearing entirely. A suppressed listing shows no Featured Offer at all: no Add to Cart button, just a See All Buying Options link that most shoppers will not click.
Across our tracked positions, those two failure modes run at roughly the same rate.
16.1%
of tracked positions are suppressed
Amazon shows no Featured Offer at all
14.8%
of tracked positions are lost
another seller holds the Featured Offer
We are not going to tell you which of those is commoner. The two sit within about a point of each other, and the ordering reverses depending on how you measure it: a snapshot of current status puts one ahead, a seven day average puts the other ahead. A gap that small is not a finding, and we would rather say so than pick whichever cut reads better.
What matters is the practical point. If you only monitor for competitors, you are watching one of two roughly equal problems. The fertilizer brand that started all of this had no competitor at all and still had a Buy Box problem.
Finding five: it is variations that change, not titles
We began snapshotting product pages daily in July and diffing them. Sellers tend to worry about their title being rewritten or their main image being swapped. Those things happen. They are not what happens most.
Variations appearing and disappearing dominate everything else, by several times the rate of title edits. A variation quietly dropping off your listing changes what a shopper can actually buy, and nothing in Seller Central announces it.
What we would tell a seller to do with this
Three things follow from the data above, and none of them require our tool to act on.
Stop trusting a national average. A single blended Buy Box percentage is the wrong shape for a metric that varies by delivery address. If your account-level number looks fine, that is not evidence that every city is fine.
Stop assuming a lost Buy Box means you were undercut. Check whether you were even the more expensive offer before you cut your price. On the evidence here, you frequently were not, and cutting price to solve a delivery speed problem just costs margin without fixing anything.
Watch for the Buy Box vanishing, not just changing hands. Suppression is about as likely as loss, and it is easier to miss because there is no competitor name to notice.
Where the numbers live
Every figure in this post is read live from the same aggregates that power our public research page, which carries the charts, the full method, and the limits. That page states plainly what this sample is not: it is the catalogue our customers chose to monitor, which means these are products somebody already suspected were worth watching. It is not a random sample of Amazon, and a seller with no Buy Box problems is under-represented in it.
The figures are free to cite with attribution. They regenerate on their own, so a number you quote today will have moved slightly by the time someone checks it, in the same way the video above has already drifted from the page it describes.
Frequently asked questions
Does the lowest price always win the Amazon Buy Box?
No. Among the products we track whose Buy Box status differs from city to city, only a minority are winning in the city where their price is lowest. More of them are winning at a higher price than the city where they are losing, and a large group shows no price difference at all. Price is one input among several, and delivery speed from nearby inventory is frequently the one that decides it.
How often does the Amazon Buy Box differ by location?
On any given day, roughly a fifth of the products we check in two or more locations show a different Buy Box status across those locations on that same day. That can mean a competitor holds it in one city and not another, or that the listing is suppressed in one place while it sells normally somewhere else.
What is Buy Box suppression and how common is it?
A suppressed listing shows no Featured Offer at all. Shoppers see a See All Buying Options link instead of an Add to Cart button. Across our tracked product-locations, suppression runs at about the same rate as losing the Buy Box to another seller. The two are close enough that we do not claim either is more common.
Can you track the Buy Box without connecting an Amazon account?
Yes. Buy Box Checker reads public product detail pages from the delivery locations you choose, so there is no Seller Central login or API connection to set up. It also means you can track any ASIN, including a competitor's, which account-linked tools cannot do.
What changes most often on an Amazon product page?
Variations being added and removed, by a wide margin. In our daily product page snapshots, variation changes run several times the rate of title edits. Most sellers watch their title and images and never see the variation churn underneath.
Check your own products
Run a free Buy Box check across several cities and see whether your own catalogue splits the way this sample does. No login, no credit card, and no Amazon account connection.
A product by Ad Badger.