The Buy Box went local. Here is what that means for Q4 pricing.

Standfirst: Amazon stopped awarding one Buy Box per listing. Q4 is when that costs the most.

Most of what you make this quarter does not happen on Black Friday.

We pulled 92 days of our own Amazon order data across last Q4 and charted it day by day. Black Friday through Cyber Monday is four days and carried 6.1% of the quarter’s orders. The 21 days from 2 December carried 30.0%. From 28 November to 22 December, orders did not drop below baseline once. Twenty five days, unbroken.

That matters for a reason that has nothing to do with deals. It means that for most of this quarter, a large volume of orders is passing through whatever pricing decision you made weeks ago. The decision does not change much. The amount riding on it changes enormously.

Which is why this is the right moment to talk about something that has changed underneath all of it.

The Buy Box went local

Amazon no longer awards one Buy Box per listing. It awards thousands, place by place.

The same listing can win the Buy Box in one location and lose it in another, on the same day, at the same price. Not a glitch, and not an edge case. It is how the allocation works now.

You may have seen this called the regional Buy Box, or described as Amazon awarding the Buy Box by zip code. Same change, different names for it.

Most sellers are still pricing as though there is one contest. Set a price, watch a national Buy Box percentage, adjust. That picture is tidy and it is no longer the thing that is happening.

As our CEO Adrian Rich put it: a repricer that sets one national price is answering a question Amazon has stopped asking.

Why one national price struggles across many Buy Box contests

If there is one contest, one price is the right shape of answer.

If there are thousands, a single national price is one answer to thousands of separate questions. It compromises across contests it cannot see. Sometimes that compromise wins. Often it wins in places where winning was cheap and loses in places where it mattered, and a national Buy Box percentage averages the two together so you never see the difference.

It is a hammer to swat a fly. You hit it sometimes, and you do a lot of damage on the way.

Not every Buy Box is worth winning

This is the part that tends to land badly at first and then make sense.

The instinct is that Buy Box share should be as high as possible. But share is bought with margin, and some contests cost more margin than the order is worth. A repricer that treats every Buy Box as equally worth winning will buy share you did not want at a price you would not have agreed to.

The question is not how many Buy Boxes you are winning. It is whether the ones you are winning are the ones worth having.

In a quarter where twenty five consecutive days run above baseline, that distinction compounds. Every day you are buying the wrong share, you are buying more of it than you would in March.

What local Buy Box awarding changes for repricing

A listing still carries one price. That has not changed and it cannot.

What changes is how that price gets chosen. Rather than pricing to a single national contest, the price can be chosen from how the Buy Box is actually being awarded across locations, accepting that one price will win in some places and lose in others, and that this is a decision rather than an accident.

That is what our engine is designed to do. Across 12 months of testing with real sellers it produced approximately 10% more profit than our previous engine.

Three things worth doing before December

Stop reading Buy Box share as a single number. A share figure that has not moved can be hiding a shift in which contests you are winning. The number is an average and averages are good at hiding things.

Decide what a Buy Box is worth to you before the volume arrives. The three weeks from the start of December are not when you want to be working out your margin floor. By then you are reacting.

Give price changes long enough to show a pattern. One contest gives you one signal. Many contests give you a distribution, and a distribution needs more than a day to read.

The quarter does not wait

The shape of Q4 is not a surprise. The run-up builds from the middle of November, holds for three weeks, and ends sharply on Christmas Eve, which is the lowest day of the whole quarter in our data.

What is different now is that the thing you are pricing into has changed, and most of the tooling in this market still reports the old picture.

Local Buy Box pricing applies to Amazon only. Local data is not currently available in the Netherlands, Belgium, Turkey, Saudi Arabia or the UAE.