Amazon has turned delivery speed into an auction. Sellers invited into its Sub Same Day network can now bid a per unit price to place additional products into it, and pay only for the units that actually ship that way. That is a different thing from a new delivery tier, and the difference is the story.
One point has to be stated plainly before anything else, because it decides whether this is news you can act on. Sub Same Day is a United States programme, it runs by invitation, and it is visible only inside Seller Central. There is nothing observable in a European storefront today, and no European seller can opt into it. What follows is a mechanism Amazon is testing in its home market, and Amazon has a long record of exporting mechanisms that work.
We are covering it now for that reason. By the time a programme like this reaches Europe, the sellers who understood the pricing logic early are the ones who bid sensibly, and the ones who meet it cold treat it as a fee rather than as a lever.
What the mechanism actually is
Sub Same Day is a separate fulfilment layer. It runs out of dedicated fulfilment centres located near 2,300 metro areas, and it offers buyers delivery of certain products in two to five hours. It was announced in March with one hour and three hour options attached.
Participation is optional. Amazon already places some products into the network automatically and without charge, selecting them itself. The change is that invited sellers can now nominate additional products by bidding: in Amazon’s own framing, “You pay only for units that actually ship through Sub Same Day, at the per-unit price you bid.”
So there are two routes into the same network. Amazon picks some of your catalogue for free, on its own criteria, and you can buy your way in for the rest. That asymmetry is worth sitting with, because it means the free placements are a signal: they tell you which of your products Amazon’s own models think will convert on speed.
There is a consumer side too. Buyers are charged for the fastest delivery windows even when they hold a Prime membership. Speed has become a paid upgrade on both sides of the transaction, which is a meaningful shift for a programme whose entire brand promise was that fast shipping is included.
From the publisher
Everything we know about running Amazon, in one course.
The commercial strategy, taught by operators who run our own brands.
Why an auction is bigger news than a new speed tier
Every previous delivery improvement Amazon has offered sellers has been a qualification. You met the criteria, or you did not, and the cost was fixed and published. An auction replaces qualification with willingness to pay, and that changes who wins.
In a qualification model, the best operator wins. In an auction, the operator with the best margin per unit wins, because they can bid higher for the same placement. Those are not the same seller. A well run business selling a low margin commodity can be outbid for speed by a worse run business selling a high margin niche product, and the buyer sees only that one arrived faster.
This is the same logic Amazon has already applied to search placement through advertising. Sponsored placement turned visibility from something you earned into something you bought, and the long run effect was that advertising cost became a permanent line in every seller’s model rather than an optional experiment. An auction for delivery speed points the same way, and sellers who watched what happened to ad spend between 2016 and today should read the parallel carefully.
There is a second order effect worth naming for anyone who shares an ASIN. Where several sellers compete on the same listing, delivery promise is already one of the things that separates them. If speed can be bought per unit, then the fastest promise on a shared listing goes to whoever bids, and the competitive question on that ASIN quietly moves from price and reliability to bid depth. That is a change in the character of the competition, not just its intensity.
Tool comparison · FastMoss vs Kalodata
When a placement becomes an auction, the winning input is data rather than instinct: which products actually convert on speed, and at what price. Category tools answer that question with very different levels of rigour. Our FastMoss and Kalodata comparison is built around the decision, not the feature list.
Comparison coming soon
FTC disclosure: EcommSphere may earn a commission if you sign up through links in this section. It does not change what we publish, and neither tool paid to be named.
The 12% claim and what is missing from it
Amazon says products placed in its Sub Same Day network have experienced 12% higher sales compared with standard FBA delivery, in areas where the programme is available. That is the number that will get quoted at you, so it is worth being precise about what it does and does not support.
What is missing is a denominator and a baseline. No published sample size, no category breakdown, no statement of how products were selected into the comparison group. The selection point is the important one: if Amazon’s own models choose which products enter the network, and those models are any good, then the products inside it were already the ones most likely to convert on speed. A lift measured on a set chosen for its likelihood to lift is not evidence that speed caused the lift.
The figure also comes with a geographic condition attached, in areas where the programme is available, which is a small phrase doing heavy work. Those areas are dense metropolitan markets, where fast delivery is both cheapest to operate and most valued by buyers. A 12% uplift there says very little about a suburban or rural catchment.
None of that makes the number false. It makes it a directional claim from an interested party about a self selected sample, which is what it should be called when it appears in a planning document. Amazon has published no base for it.
Tool comparison · FastMoss vs Kalodata
Bidding for placement only works if you know what a unit of speed is worth in your category. That is a margin question and a demand question at the same time. We are running FastMoss and Kalodata through the same brief so the trade-offs are visible before you commit.
Comparison coming soon
FTC disclosure: EcommSphere may earn a commission if you sign up through links in this section. It does not change what we publish, and neither tool paid to be named.
What European sellers should take from this
Nothing operational, and something strategic. Operationally there is no action available: the programme is not open to you, and no amount of preparation changes that this quarter.
Strategically there are three things worth carrying. The first is that Amazon is willing to charge buyers for speed even inside Prime, which tells you the economics of same day do not close on subscription revenue alone. The second is that speed is becoming a variable cost you bid for rather than a service level you inherit, and variable costs need a model. The third is that Amazon is building dedicated infrastructure for this rather than layering it onto existing fulfilment centres, and dedicated infrastructure is a commitment, not a test.
The European timing question is genuinely open. Amazon’s European network is denser in some metropolitan areas than its United States equivalent and much thinner elsewhere, and same day economics depend almost entirely on density. A version of this arriving in London, Paris, Milan and Berlin first would be unsurprising. A pan European rollout would be.
One more thing about the 2,300 metro areas figure, because it is easy to misread as coverage. It describes where Amazon has put dedicated fulfilment capacity near, not where two hour delivery is offered on your products. Network reach and product eligibility are separate constraints, and the second one is much tighter than the first. A programme can be near almost everyone and available for almost nothing.
What to do now
- Work out which of your products would plausibly convert on a two hour promise rather than a two day one. In most catalogues it is a small, identifiable group: replenishables, gifts, and anything bought under time pressure.
- Know your contribution margin per unit at SKU level, not at category level. An auction is unbiddable without it.
- Watch which of your products Amazon places into fast networks for free in any market where it does so. That selection is free information about what its models believe.
- Do not build a European plan on this yet. There is no European programme to plan against, and a plan built on an assumed launch date is a plan built on nothing.
What we are not saying
We are not saying European sellers can use Sub Same Day, because they cannot: it is United States first, by invitation, and visible only inside Seller Central. We are not endorsing the 12% figure, which Amazon has published without a base. What we are saying is that Amazon has started auctioning delivery speed by the unit, that auctions reward margin rather than operational quality, and that this is the mechanism worth understanding before it arrives anywhere else.
Sources
- EcommerceBytes, https://www.ecommercebytes.com/2026/08/29/amazon-lets-sellers-pay-extra-for-ssd-sub-same-day-shipping/