For about a decade, used books on Amazon ran on one simple rule. Send your books to FBA, price them a bit above the merchant-fulfilled offers, and the Prime badge wins you the Buy Box anyway. Everyone called it the prime bump. A book sitting at $8 from an FBM seller could sell for $12 in FBA, and nobody thought twice.
That rule is gone. Since November 3, 2025, the used Buy Box for books and media on Amazon.com no longer favors Amazon fulfillment. I follow several bookseller communities, including one with 37,000 members, and I've read close to two years of threads about this. My takeaway fits in one line: the Buy Box didn't die, the rules changed, and the sellers still standing are the ones who learned the new ones.
Here's what happened, what it broke, and the pricing and FBM vs FBA rules sellers actually use now.
What happened: two dress rehearsals, then the real thing
Amazon didn't flip the switch in one go. It tested the change twice before making it permanent, and never warned sellers in advance.
| When | What sellers saw | Reported FBA sales drop |
|---|---|---|
| Feb 24 → Mar 15, 2025 | Used Buy Box moves to big FBM sellers (ThriftBooks, World of Books, Goodwill), even when the FBA offer is cheaper. Rolled back on the evening of March 15. | −50% to −95% |
| Sept 16 → 30, 2025 | Second shift to the lowest FBM offer. Two or three used Buy Boxes at once, display varies by buyer account. | −43% to −85% |
| Nov 3, 2025 → today | Permanent. Condition comes first, then price, then a little delivery speed. | −25% to −97% |
| July 7, 2026 | Featured Offer eligibility criteria removed, per sellers. | — |
On November 12, 2025, an Amazon moderator on the seller forums wrote that the featured offer selection is "agnostic of the fulfillment channel (AFN or MFN)." On November 24, an Amazon message shared by several sellers said the changes "enable MFN sellers to directly compete against AFN in Used media." (MFN is FBM, AFN is FBA.)
So it's intentional. A 20-year FBM veteran has a rule for every Amazon change: if it hurts for more than four weeks, assume it's permanent. We're close to a year in.
For what it's worth, UK sellers say they went through the same thing around 2023. One of them told US sellers in early 2026: "We're 2 years ahead of you guys." Outside textbook season, he said, almost nobody sells books through FBA in the UK anymore.
The damage, in numbers
These are self-reported numbers from sellers looking at their own accounts, not Amazon data. They all point the same way.
| What sellers reported | Before | After Nov 3, 2025 |
|---|---|---|
| Share of listings where one seller saw FBA holding the used Buy Box | over 70% | about 20% |
| One seller's scan of roughly 107,000 listings | — | 19% FBA Buy Box |
| A 100-book sample tracked over several weeks | — | FBM won in 85%, then 90%, then 94% |
| One-week FBA sales change (two sellers) | — | −84% and −97% |
| FBA inventory of one seller who switched | 4,000 books | 500 books |
Experienced people quit. One FBA seller said he was "working for $7 per hour now." The line you hear most since summer 2026 is that FBA-books-only isn't a viable business anymore.
The flip side: high-ASP FBM sellers did fine. Some reported +59% during the first episode. The Buy Box didn't vanish. It changed hands.
The hidden trap: your tools show a Buy Box that doesn't exist
This is the part fewer people talk about, and it's probably the most expensive.
Barcode scouting apps and repricers pull their Buy Box data from Amazon's API. According to a lot of sellers, since October–November 2025 that API still returns the old Buy Box, the one where FBA wins. A large bulk seller put it bluntly this spring: no third-party software has had fully accurate API data since the logic change. Not scouting apps, not repricers, not listers.
In practice:
- Bad buys. Your app says the FBA Buy Box is $15. The real Buy Box is a $6 FBM offer. You just bought a dud.
- Repricers chasing a ghost. "Match Buy Box" matches nothing. As one seller put it, match buybox doesn't match the buybox.
- Phantom sales. During the September 2025 episode, sellers saw a single copy "sell" multiple times, up to 56 orders for one book.
Keepa, which reads product pages instead of the API, was the only place showing the real used Buy Box. A few scanners only started patching this in mid-2026, so check what yours actually reads before you trust it.
Then came DD+7. Announced on September 29, 2025, it hit older accounts on March 12, 2026: you get paid seven days after delivery. Sellers went one to three weeks with no payout, and in June 2026 some were still receiving only 20–30% of what they were owed. Fewer FBA sales plus frozen cash at the same time is what pushed a lot of people out.
How the used Buy Box is won now
Sellers spent a year rebuilding the logic by trial and error. Current consensus:
- Condition first. Very Good beats Good as long as it stays reasonably close, usually quoted at 10–20% above. Acceptable almost never wins.
- Price second, compared against offers in the same condition, FBM included.
- Delivery speed a little, much less than before.
A study a seller shared in early 2026 put a number on the Like New premium: Amazon seems to value Like New only about 6% over Very Good. So if you list Like New, you have to stay very close to the Very Good price to win.
That kicked off an over-grading war. Some sellers now list everything as Like New. One veteran's summary: "Like New now means 'still has pages.'" Bad idea long term: returns, a higher refund rate, Buy Box suppression, and occasionally suspensions. Grade honestly.
Pricing rules that work in 2026
Here's what the sellers who are still profitable actually do:
| Situation | Rule | Why |
|---|---|---|
| Reference price | Lowest offer in the same condition, FBM included | The Buy Box no longer just compares FBA offers |
| Very Good or Like New | Lowest Good + 10–20% | Better condition wins if the gap stays reasonable |
| FBA offer | No more than ~10% over the lowest FBM | Past that, Prime doesn't save you |
| Acceptable copies | Don't buy them for Amazon, or sell elsewhere | They almost never get the Buy Box |
| Price wars | Don't undercut by a penny | "If you undercut by a penny in a week you will have books that are worthless" |
| Sourcing | Judge a book on the lowest current and 1-year price, any condition or channel | That's its real floor |
One old-world rule to retire: a book at $5–6 FBM and $15 FBA used to be a buy thanks to the prime bump. Sellers estimated the bump at roughly 2x under 200k rank, 1.5x around 500k, 1.25x around 1M and 1.1x around 1.5M. Today that book is a dud.
FBM vs FBA: the decision grid
Sellers didn't all go 100% FBM. Most now sort books into two buckets.
| Factor | Lean FBA | Lean FBM + other channels |
|---|---|---|
| Sales speed | Fast (rank under 500k–1M) | Slow, long tail |
| Price | At least $15–20, stable value | Over $20, especially pricey, rare, niche |
| Condition | Good to Very Good | Any, with an accurate description |
| Format | Standard books, DVDs and CDs (low FBA fees vs Media Mail) | Oversize, heavy, art books |
| Channels | Amazon only | Amazon + eBay, AbeBooks, Alibris, Biblio |
Some numbers to calibrate: one seller reports $6 average profit FBM vs $3 FBA; a large bulk seller now aims for about 2% of inventory in FBA. On the other side, one seller still sells twelve times more in FBA than FBM. Both models work. The sorting is what matters.
Watch the velocity gap. FBA typically sells 0.5–1% of inventory per day; FBM more like 0.25–0.5%. You need more books listed in FBM for the same revenue, which is why cross-listing to eBay became almost mandatory. Seller-shared rules for per-book decisions: FBM above roughly 4M rank, at least 6 sales in 12 months and $18–20 to bother with FBA, and liquidate FBA stock worth under $20 rather than pay to pull it back.
Fees still decide a lot. Amazon takes a 15% referral fee plus a $1.80 closing fee on books, and several sellers put total FBA fees at 45–55% of the sale price. Before you send a box in, run the book through the BiblioScan FBA calculator. On a $10 book, the math is usually ugly.
How to check the real Buy Box before you buy or price
The method sellers recommend:
- Don't trust the Buy Box in your scanning app unless the app says explicitly that it reads the real used Buy Box. BiblioScan now does, more on that below.
- Open Keepa. Look at the used offers and price history. When offer count and rank drop together, that's a real sale.
- Compare within the same condition. Lowest Good, lowest Very Good, FBA and FBM together.
- Look at the product page yourself, ideally from a Prime and a non-Prime account. In September 2025 the Buy Box differed by buyer.
- Read sales history, not current rank. A single rank snapshot tells you nothing about how often a book sells.
Where BiblioScan fits (and where it doesn't)
Let me be straight. BiblioScan is not a repricer: it won't move your prices for you. But you can now check the real used Buy Box right inside BiblioScan, not the old FBA-favoring one the API keeps returning.
Open any book and tap See the Buy Box. A couple of seconds later you get the used Buy Box as it stands: total price with shipping, the winning condition, and whether it's FBA or FBM. Below that sits what the new rules made essential, the last 90 days: how often FBA actually held the Buy Box, which condition wins most, the sellers who held it longest and at what average price, and the latest changes of winner. The data comes from Keepa's Buy Box tracking, not from the API that's stuck on the old rules.
It's on demand, because each check pulls fresh data: 2 barcode credits per book, refunded if there's nothing to show. The result stays saved on the book, so you can come back to it for free, with the time it was taken. It's a snapshot, though. The Buy Box moves during the day and can differ from one buyer account to another, so step 4 above still applies on the books that matter.
The rest of BiblioScan answers the question that matters now that the prime bump is gone: should I buy this book, and where should it go? You take one photo of a shelf, a stack or an open box, and the AI recognizes every book by its spine. For each one you get the median price it actually sold for, how many times it sold, a recommended max buy price, and BiblioRank to put the best books at the top of the pile. A compass, not a guarantee.
Three things in particular line up with the new rules:
- Sales history, with the Keepa chart built in. Price and sales-rank history sit right under every result, so step 5 of the check above happens without leaving the scan.
- Net profit for FBA or FBM. Connect your Amazon Seller account and the net profit after commission and fees is computed for your fulfillment mode. You also see whether you're approved to sell the book, and whether you already have it in stock.
- A cash floor for the duds. Every US scan pulls live offers from 20+ buyback sites (BooksRun, TopDollar4Books, World of Books and more), and one tap takes you to recent eBay sales. The books that no longer make sense in FBA still have somewhere to go.
Want to try it on a single title first? The free ISBN lookup runs without an account. For the books that don't pass, here's my rundown of the best book buyback sites. And if you're new to all this, start with how to scan books for resale and where Amazon really sits in the US used book market.
Bottom Line
Since November 3, 2025, Amazon.com's used book Buy Box no longer favors FBA, after two test runs in 2025. Sellers report FBA sales down 25–97%, and tools fed by the API still show a Buy Box that doesn't exist. The Buy Box is now won on condition first, then price within that condition. Price against the lowest same-condition offer, FBM included; go lowest Good + 10–20% for Very Good and Like New; skip Acceptable; keep FBA for fast books with stable value and media; move the long tail and anything over $20 to FBM and cross-list it. Trust sales history and what BiblioScan shows you over the Buy Box a typical scouting app displays, and when a book matters, pull up its real used Buy Box right in BiblioScan. Sort your next box with 5 free credits, no card required.
FAQ
Does FBA still win the Buy Box on used books?
Much less than before. Since November 3, 2025, Amazon's used featured offer selection on Amazon.com is fulfillment-agnostic, and sellers report FBA holding about 20% of used Buy Boxes, down from over 70%.
How do I win the used Buy Box on Amazon?
Condition first, then price within the same condition, then a little delivery speed. Very Good beats Good if it's within 10–20% on price. Acceptable almost never wins.
Should I sell used books FBA or FBM in 2026?
FBA for fast-selling books with stable value at $15–20 or more, plus DVDs and CDs. FBM with cross-listing for the long tail, slow sellers and books over $20.
Why is my scanner showing the wrong Buy Box?
Sellers report that Amazon's API has kept returning the old FBA-favoring Buy Box since late 2025, and most scanners read that API. BiblioScan shows the real used Buy Box on demand, from Keepa's Buy Box tracking: current price, condition, FBA or FBM, and how often FBA actually won it over the last 90 days. Either way, price against the lowest offer in the same condition, FBM included.
Should I list everything as Like New to win the Buy Box?
No. The advantage is reportedly only about 6%, and over-grading brings returns, refunds and suspension risk. Grade honestly.
Will the FBA Buy Box advantage come back?
Plan as if it won't. Amazon called the change intentional, it's been in place for almost a year, and UK sellers say their market has run this way since around 2023.