Something is shifting inside Amazon’s Buy Box engine, and the seller community is starting to notice. Over the past six weeks, a growing number of high-volume third-party sellers have reported unexplained Buy Box suppression on listings where they’ve historically held dominant position — even while maintaining competitive pricing and strong seller metrics. Sources close to the matter say the anomalies aren’t bugs. They’re signals.
According to two individuals with direct knowledge of Amazon’s Marketplace development roadmap — neither of whom was authorized to speak publicly — the company is reportedly running a phased A/B test on a revised Buy Box scoring model that places significantly less weight on landed price and substantially more weight on what Amazon is internally calling “delivery confidence scoring,” a composite metric that allegedly factors in carrier reliability, warehouse proximity to the buyer, and real-time inventory depth. The test is said to be live in at least three product categories: home goods, sporting goods, and pet supplies.
Amazon has not confirmed any algorithm change, and a spokesperson declined to comment specifically on Buy Box methodology. But the circumstantial evidence in seller forums and aggregator back-offices is piling up fast.
What Are Sellers Actually Seeing in Their Dashboards?
The pattern is consistent enough to be alarming. Sellers are reporting Buy Box loss on ASINs where they’re priced 3–5% below the next competitor, but where their inventory is sitting in a single FBA fulfillment center rather than distributed across Amazon’s network. Meanwhile, sellers with Inventory Placement Service enabled and multi-node distribution are reportedly holding Buy Box at higher price points than would have been competitive six months ago.
“We noticed it first around May 12th,” said Kevin Pak, Director of Marketplace Strategy at Pattern, the commerce acceleration platform that manages over $4 billion in annual Amazon GMV for enterprise brands. “We had five brands lose meaningful Buy Box percentage within the same 72-hour window. No pricing changes, no metric degradation. The only common thread was single-node inventory concentration.”
“If this scoring model goes full rollout, the replenishment playbook for every mid-market seller in America changes overnight. You can’t win on price alone anymore — you have to win on fulfillment geography.” — Kevin Pak, Pattern
Pattern has reportedly pulled internal data across 200+ brand accounts and identified a statistically significant correlation between FBA inventory distribution breadth and Buy Box retention rates since mid-May. The company is allegedly preparing a seller advisory for its brand clients but is waiting for more data before going public.
Is Amazon Deliberately Penalizing Single-Node FBA Sellers?
That’s the question rattling seller Slack groups and aggregator finance teams. The unconfirmed hypothesis making the rounds is that Amazon is using the Buy Box mechanism — rather than explicit policy — to pressure sellers into adopting its Inventory Placement Service and distributed replenishment models. The incentive structure is obvious: distributed inventory reduces Amazon’s internal transfer costs and improves its Prime delivery promise performance.
“Amazon doesn’t need to change its terms of service to change seller behavior,” said Dani Avitz, co-founder of Incrementum Digital, an Amazon-specialized agency managing roughly $180 million in annual ad spend. “They can just quietly shift the algorithm and let the data do the work. Sellers will adapt or lose revenue. It’s a very Amazon way to enforce infrastructure compliance without a policy memo.”
“The Buy Box has always been a behavior modification tool dressed up as a customer experience feature. This would just be the latest version of that.” — Dani Avitz, Incrementum Digital
Avitz says his agency has observed Buy Box win rate drops of 8–14 percentage points on affected ASINs for several mid-size clients, translating to estimated revenue impacts of $40,000 to $120,000 per month per brand at current traffic volumes. He described the losses as “not catastrophic yet, but trending in a direction that demands action.”
How Are Aggregators Responding to the Reported Shift?
Amazon aggregators — already under considerable financial pressure following the 2023–2024 consolidation wave — are reportedly among the most exposed to this alleged change. Several aggregator operations run lean inventory models that rely on centralized FBA stocking to preserve cash flow. If delivery confidence scoring is real and scales, the capital requirements for competitive Buy Box positioning could increase substantially.
Sources familiar with operations at Thrasio — which has been quietly rebuilding its brand portfolio under CEO Greg Greeley since its restructuring — say internal teams have been tasked with auditing every active ASIN for inventory distribution exposure. The audit, reportedly launched in late May, is described as “urgent” in internal communications. Thrasio declined to comment.
At Perch, which operates roughly 70 Amazon brands following its own restructuring, sources say the inventory ops team has been in weekly discussions with their FBA account manager about the observed anomalies. One source described the internal tone as “cautiously alarmed” — acknowledging something is changing but not yet certain of the mechanism.
- Thrasio: Reportedly conducting an emergency ASIN-level inventory distribution audit across its active brand portfolio.
- Perch: Said to be in active dialogue with Amazon account managers about Buy Box anomalies observed since mid-May.
- Pattern: Has internally correlated inventory node count with Buy Box retention across 200+ brand accounts.
- Incrementum Digital: Tracking 8–14 percentage point Buy Box declines on affected client ASINs.
What Does This Mean for FBM Sellers Competing for the Buy Box?
Interestingly, the reported algorithm shift may create a narrow window of opportunity for Fulfillment by Merchant sellers who operate distributed warehouse networks — particularly those using third-party logistics providers with multi-node footprints. If delivery confidence scoring genuinely rewards proximity and reliability over price, an FBM seller using a 3PL with same-day or next-day coverage in major DMAs could theoretically compete more effectively than a single-node FBA seller.
“We’ve had two FBM clients actually improve their Buy Box percentage in the last month while their FBA competitors degraded,” said Avitz. “That’s almost unheard of historically. It’s anecdotal, but it lines up with the theory.”
Tools like Jungle Scout and Helium 10 have not yet updated their Buy Box tracking dashboards to surface inventory distribution as a variable, according to multiple sellers who contacted Ecommerce Times. Both companies confirmed they are “monitoring algorithm signals” but declined to confirm whether product updates are in development in response to the reported change.
Could Amazon Be Preparing a Paid Tier for Buy Box Distribution Priority?
Perhaps the most provocative rumor circulating in the ecosystem — and it is explicitly unconfirmed — is that Amazon’s alleged scoring revision is a precursor to a paid program that would allow sellers to purchase enhanced “delivery confidence” status through a premium inventory placement tier. Think of it as a Buy Box insurance product layered on top of standard FBA.
One source, described only as a former Amazon Marketplace category manager who left the company in early 2026, told Ecommerce Times that internal conversations about monetizing fulfillment quality signals have been ongoing since at least 2024. “The idea of tiered Buy Box access based on fulfillment investment has been floated in Marketplace economics conversations more than once,” the source said, speaking on condition of anonymity. “Whether it ever becomes a formal product is a different question. But the directional intent — extracting more infrastructure spend from sellers — is very consistent with where the business model has been heading.”
“Every time Amazon has a lever that drives seller infrastructure spend, they eventually find a way to formalize it. The question is always timing and packaging.” — Former Amazon Marketplace category manager, speaking anonymously
Amazon has historically been resistant to framing any Buy Box mechanic as pay-to-play, given regulatory scrutiny in the EU and ongoing FTC attention in the U.S. A formalized paid tier would almost certainly attract immediate antitrust attention. But a de facto scoring advantage embedded in an opaque algorithm is considerably harder to challenge legally — which may be precisely the point.
What Should Sellers Do Right Now?
Operationally, the consensus among agency leads and aggregator operators who spoke with Ecommerce Times is clear, even if the algorithm change itself remains unconfirmed: sellers should not wait for Amazon to announce anything.
- Audit inventory distribution immediately: Pull your top 20 revenue ASINs and map which FBA fulfillment centers currently hold your stock. Flag any ASIN concentrated in a single node.
- Enable or expand Inventory Placement Service: If you’re currently using the default inventory placement, consider switching to distributed placement for your highest-revenue SKUs, even at the added cost.
- Monitor Buy Box win rate daily: Set up alerts in Seller Central or through third-party tools like Helium 10 or Jungle Scout for any Buy Box percentage drop exceeding 5 points in a 7-day window.
- Test FBM on select ASINs: If you have access to a distributed 3PL, running a parallel FBM offer on affected listings may reveal whether the delivery confidence hypothesis holds in your category.
- Document anomalies for your account manager: If you have a dedicated Amazon account manager, escalate with specific ASIN data. Multiple escalations from different sellers may accelerate any official acknowledgment from Amazon.
The broader implication is one that sophisticated marketplace operators have long understood but mid-market sellers are still internalizing: Amazon’s algorithm is a business model instrument, not a neutral ranking system. When Amazon’s infrastructure economics and third-party seller behavior diverge, the algorithm tends to converge them — quietly, gradually, and without a press release. This may be the latest iteration of that dynamic, and the sellers who move first will be the ones who hold position when the dust settles.
Ecommerce Times will continue to track this story as more seller data becomes available and as Amazon’s testing window — assuming it exists — moves toward any potential broader rollout.
