Amazon Fresh is not a hybrid channel anymore. In late January 2026, Amazon announced it was closing every Amazon Fresh store and every Amazon Go store it operated, 57 Fresh locations and 15 Go locations, 72 doors total, with most of them shut by February 1 and California locations given a 45-day extension under state notification law. I've been operating on Amazon since 2016, across three different grocery cycles, and I have not seen a channel change this fast without a single adjustment to the number brands are still expected to hit.
The stores are closed. The forecast is not. If a real share of a brand's Fresh volume used to walk out through a physical door, Amazon still wants that volume this year, it just wants it delivered through a search bar instead. Most Fresh vendors are not built for that swap. Here is exactly what changed, what Amazon has and hasn't adjusted for it, and what it actually costs a brand that gets this wrong.
Amazon closed all 57 Amazon Fresh stores and all 15 Amazon Go stores, 72 physical locations, and folded the entire Fresh program into pure online delivery. Amazon's own announcement framed the move as concentrating investment in Whole Foods Market and same-day delivery infrastructure instead of the physical Fresh and Go format. Most locations closed February 1, with California stores held open an additional 45 days to comply with state labor notification requirements.
Amazon shut all 57 Amazon Fresh and 15 Amazon Go stores in early 2026, most of them by February 1.
Fortune's coverage of the shutdown landed on the same conclusion most operators had already reached watching the format from the outside: the physical Fresh and Go stores were expensive to run relative to what they generated, and Amazon didn't need the square footage to keep winning the category, it needed the delivery network. For a shopper, that's a convenience story. For a vendor who spent years earning placement in those 72 stores, it's a channel that vanished overnight with a wind-down period shorter than two months.
The digital side of the business did not slow down to match. Amazon's same-day perishable grocery delivery now reaches more than 2,300 U.S. cities and towns, and the company has reported perishable grocery sales growing by a factor of 30 or more since January 2025. Whichever exact multiple you use, the direction isn't in question: Amazon is routing more grocery volume than ever through delivery, from a smaller physical footprint than it's had in years.
For a Fresh vendor, that's the whole story compressed into one sentence: the shelf you used to negotiate for is gone, and the shelf you didn't build yet is the only one left.
Because the planning cycle that sets your purchase order volume and your Annual Vendor Negotiation targets looks at trailing performance and category growth assumptions, not at how much of your prior volume came through a channel that no longer exists. Nobody at Amazon issues a memo re-basing your forecast because 72 stores closed. The number holds. The channel that used to help you hit it does not.
This isn't the same conversation as a third-party seller adjusting to a marketplace shift. Fresh vendors sell 1P, through Vendor Central: Amazon buys the inventory, controls much of the retail experience, and negotiates fees, marketing contributions, and cost terms annually through the Vendor Negotiation process. When the retail environment changes this much, a 3P seller can react in days. A 1P vendor is locked into a purchase-order and negotiation cycle that assumes the channel still looks the way it looked when the number was set.
I've sat on the brand side of enough of those negotiations to know exactly how that math gets built, and it does not have a line item for "and then Amazon closed its own stores." If a brand historically ran half its Fresh volume in-store, that volume doesn't disappear from the target, it just has to be found somewhere else, digitally, inside the same calendar year.
The volume that used to come through the store does not leave the forecast. It becomes a gap you have to close digitally, inside the same year.
Probably more than the topline number suggests, and the two blindest spots are zip-code coverage and distribution-center placement. Most Amazon Fresh vendors are live in well under the full set of zip codes where Fresh delivery is actually available, and most vendor catalogs are stocked in only a handful of Amazon's Fresh distribution centers rather than the full network.
That matters because "the shelf is digital now" doesn't mean "the shelf is everywhere now." A brand can fix every piece of content and run every ad campaign correctly and still be functionally invisible to a large share of Fresh-eligible households, not because the algorithm doesn't like the brand, but because Amazon hasn't routed inventory into the distribution centers that serve those households yet. Expansion into new zip codes and new DCs is now doing the job that opening a new store used to do. Most brands aren't managing it as a discipline. They're waiting for it to happen.
Most vendors are live in only a slice of eligible zip codes and stocked in only a few distribution centers, so a large share of Fresh households never sees them.
| Old lever | Digital equivalent |
|---|---|
| New store openings | New Fresh distribution center placement |
| Regional distributor relationships | Zip-code-level fulfillment mapping |
| Broker's local trade relationships | Amazon's own DC allocation decisions |
| Store count | Eligible zip-code coverage |
It means every job a physical shelf used to do, get noticed, make the sale, keep the customer coming back, now has a digital equivalent, and none of it happens automatically.
Discovery used to be shelf placement and slotting: whoever negotiated the best spot got seen first. Now discovery is search placement and sponsored advertising, the ad is the slotting fee, and it renews every day instead of once a year.
Merchandising used to be packaging: a shopper picked the box up, read the label, made a decision in the aisle. Now merchandising is the image gallery, the bullets, and the A+ Content, the only pitch a product gets in the three to five seconds before a shopper's thumb moves to the next listing.
Loyalty used to be habit: a shopper walked the same aisle every week and picked up the same box out of routine. Now loyalty is Subscribe and Save, the only mechanism that manufactures that same habit on a screen instead of in an aisle.
Discovery, merchandising, and loyalty did not disappear. Each one moved to a digital mechanism a brand has to build on purpose.
None of this is a hypothetical. It's the operating reality for every brand still selling into Amazon Fresh in 2026, whether or not anyone at the company has said it out loud yet. Part 2 of this series covers exactly where the relationship that got most brands into Amazon Fresh in the first place, the traditional CPG or grocery broker, runs out of road on a fully digital channel, and why that isn't a broker failure. Part 3 covers the actual playbook for closing the gap, and Part 4 covers who runs that playbook and how.
If you want a straight read on where your account stands against these numbers first, TopRank runs a no-cost Vendor Audit on Amazon Fresh accounts: read-only Vendor Central access, five to seven business days, and the findings are yours to keep whether or not we ever work together.
A no-cost Vendor Audit: content, advertising, distribution coverage, and chargeback exposure, reviewed by a senior operator. Read-only access. Findings are yours to keep either way.
Request the Vendor Audit