Amazon has cut fewer than 1,000 corporate roles, with the reductions falling mainly on its Stores division, the organisation that runs the company’s core retail and e-commerce business. Reuters reported the cuts late on October 7, and The Information reported that hundreds of Stores employees had been let go. Business Insider put the figure below 1,000 across several teams.
The affected staff sit in the United States, the United Kingdom and India, according to reports citing people familiar with the decision. Posts circulated on Amazon’s internal Slack channels pointed to customer service and selling partner services as among the teams hit, with smaller reductions elsewhere in the company.
An Amazon spokesperson said the company had “adjusted parts of our Stores business because we believe this structure will better enable us to deliver on our priorities.” Amazon did not publish a memo or a headline number, which marks a change in approach from the two large, announced rounds that preceded it.
The timing is what makes this more than a routine reorganisation. The cuts landed within hours of the close of Prime Big Deal Days, Amazon’s 48-hour pre-holiday sales event, and three weeks before the company reports third-quarter results on October 29.
In short
- Scale: fewer than 1,000 corporate roles, described by one report as hundreds in the Stores division alone.
- Where: the Stores division took the bulk, with customer service and selling partner services named in internal posts; staff in the US, UK and India are affected.
- Context: this follows roughly 30,000 corporate cuts announced in two tranches, 14,000 in October 2025 and 16,000 in January 2026.
- Framing: Amazon issued a short statement rather than a memo, and gave no company-wide number, unlike the earlier rounds.
- Next marker: Q3 2026 results on October 29, where guidance of $197bn to $202bn in net sales meets a 2026 capital spending plan of about $220bn.
What exactly did Amazon cut on October 8?
The reductions are corporate, not warehouse. Nothing in the reporting points to fulfilment centre or delivery station staffing, which Amazon normally expands rather than contracts in the weeks before the holiday peak. The people losing roles are salaried employees in retail, support and operations teams.
The headline number is unusually soft. Reuters and Business Insider both framed the total as fewer than 1,000 roles across teams. The Information described hundreds of layoffs specifically in Stores. Those two statements are compatible: a Stores-centred action of several hundred, plus scattered reductions elsewhere, lands under 1,000 in aggregate.
Amazon has not confirmed a figure. That matters when comparing this action with the two previous rounds, both of which came with a precise, company-authored number and a signed memo. The absence of either is a deliberate choice about how the company wants this read.
The reporting chain itself is informative. The Information published first, late on October 7 US time, with a Stores-specific framing. Reuters followed within the hour with a broader “mainly retail division” account, and Business Insider supplied the sub-1,000 aggregate. Indian and Singaporean outlets picked it up through the Asian morning.
No outlet reported a memo. In both earlier rounds, the document came first and the coverage followed. Here the sequence ran the other way: reporters found the cuts, then Amazon supplied a sentence. That ordering is the clearest evidence that this was not meant to be an announcement.
The teams named so far
Two teams recur in the reporting. The first is customer service, the organisation that handles buyer contacts across chat, phone and email. The second is selling partner services, the function that supports third-party merchants on the marketplace, from account health to listing problems and appeals.
Both names come from internal Slack posts described to reporters rather than from a company disclosure, so they should be treated as reported rather than confirmed. Neither is a peripheral function. Customer service and seller support are the two places where Amazon’s retail business touches people directly when something has gone wrong.
That is the detail worth holding onto. A cut to a speculative hardware project tells you about Amazon’s appetite for risk. A cut to buyer and seller support tells you about Amazon’s confidence that it can absorb the same volume of contacts with fewer people, which is a claim about tooling, automation or service expectations.
Which countries are affected
The United States, the United Kingdom and India all appear in the reporting. India stands out, because several Indian outlets led with the local angle: Amazon’s Indian retail staff are among those affected, in a market where the company has spent years building out a marketplace under shifting rules.
India has also been the source of a steady stream of compliance work for Amazon this year, including the certification sweep that saw Amazon and Flipkart pull uncertified CCTV listings after India’s Bureau of Indian Standards rules began to bite. Support headcount in that market carries a regulatory load as well as a commercial one.
The UK’s inclusion is a reminder that Amazon’s European retail organisation runs on the same functional lines as the US one. When a global team is restructured, the cut tends to show up in every country where that team has seats, rather than in one head office.
Why does the timing, right at the end of Prime Big Deal Days, matter?
Amazon ran Prime Big Deal Days on October 6 and 7, a 48-hour Prime-exclusive event across 22 countries designed to pull holiday demand forward into early October. The layoffs were reported as that event closed.
Companies usually avoid cutting staff during a peak trading window, for two reasons. The first is operational: a sales event is exactly when buyer contacts and seller escalations spike. The second is reputational: announcing reductions while asking the same organisation to deliver a record event invites an obvious question about priorities.
There are two readings of the choice. One is that the cuts were planned for the start of the fourth quarter and the event calendar simply collided with them. The other is that Amazon deliberately waited until the event was complete, then moved before the quiet period ahead of earnings on October 29.
Both readings point the same way on substance. Whatever drove the calendar, Amazon has decided that its retail support organisation can enter the holiday quarter smaller than it left the third quarter. That is a forecast about contact volumes and about the tools handling them.
How does this fit the 30,000 corporate cuts since October 2025?
Amazon’s current restructuring began in late October 2025 and has run through two announced tranches. Taken together they form the largest workforce reduction in the company’s history, above the roughly 27,000 roles cut in 2023.
The October 2025 tranche
On October 28, 2025, Amazon said it would cut about 14,000 corporate roles. Beth Galetti, the senior vice president for people experience and technology, wrote that the company needed to be “organized more leanly, with fewer layers and more ownership, to move as quickly as possible for our customers and businesses.”
That framing was structural rather than financial. The stated problem was layers of management and slow decision-making, not revenue. It fit a wider programme Amazon had been running for two years to reduce the ratio of managers to individual contributors.
The January 2026 tranche
On January 28, 2026, Amazon confirmed a further 16,000 corporate roles globally, taking the total since October 2025 to roughly 30,000. Coverage at the time described it as an anti-bureaucracy push with an artificial intelligence dimension, and as the largest single reduction Amazon had ever made.
Galetti did not rule out further cuts, but pushed back on the idea that Amazon was settling into a pattern. She said the company was not trying to create “a new rhythm” of broad layoffs every few months, adding: “That’s not our plan.” She also said every team would continue to evaluate ownership, speed and capacity to invent, and make adjustments as appropriate.
How it compares with the 2023 round
Amazon’s previous restructuring cycle removed roughly 27,000 roles across 2022 and 2023, concentrated in devices, Alexa, advertising and human resources. The current cycle is larger and differently targeted, reaching into the retail organisation itself rather than the adjacent bets.
That shift in target is the substantive change between cycles. In 2023, Amazon cut the parts of the company that were not yet paying for themselves. In 2025 and 2026, it has been cutting coordination inside the business that generates most of its revenue.
| Round | Date announced | Roles | Scope | How Amazon framed it |
|---|---|---|---|---|
| First tranche | October 28, 2025 | About 14,000 | Corporate, company-wide | Memo from Beth Galetti: fewer layers, more ownership |
| Second tranche | January 28, 2026 | About 16,000 | Corporate, global | Anti-bureaucracy push; largest cut in company history |
| This action | October 7 to 8, 2026 | Fewer than 1,000 | Mainly Stores division; US, UK, India | Short spokesperson statement: structure “will better enable us to deliver on our priorities” |
The scale gap is the point. This round is roughly 3% of the January tranche. It is not a third mass layoff, and treating it as one misreads the signal.
Does this break the “no new rhythm” pledge?
On a literal reading, no. Galetti’s January comment was about avoiding broad layoffs every few months. A sub-1,000 action nine months later, confined mainly to one division, is not a broad layoff, and the gap is not a few months.
On a practical reading, the distinction is thinner than it looks. Galetti’s second sentence, that every team would keep evaluating and adjusting, described exactly this kind of event: periodic, unannounced, team-level reductions that never add up to a company-wide number.
That is the structural shift worth naming. Amazon appears to be moving from announced mass layoffs with a memo and a figure, to continuous trimming at divisional level with a one-sentence statement. The second mode is harder to track, harder to aggregate and much less visible in headline counts.
For anyone trying to measure Amazon’s corporate headcount trajectory, that is a genuine measurement problem. The company’s reported employee total is dominated by its operations workforce, so corporate reductions of this size do not move it. Without a disclosed figure, the only record is reporting.
What do the cuts say about the Stores business?
Stores is Amazon’s retail engine: first-party selling, the third-party marketplace, the Prime shopping experience and the physical grocery and convenience estate. It is also the part of Amazon that competes directly with Walmart, Target and Temu rather than with Microsoft and Google.
Customer service and the automation question
Customer service is the most automatable large function in retail. Amazon has spent years pushing self-service returns, automated refunds and conversational assistance into the buyer flow, with the explicit aim of reducing contacts that need a human.
A reduction in that team is consistent with Amazon judging that the deflection rate is now high enough to carry the holiday quarter. It is also the kind of judgment that gets tested in public in December, when delivery exceptions peak.
The risk is asymmetric. If the tooling holds, nobody notices. If it does not, the failure shows up as unanswered contacts during the highest-volume weeks of the year, at a moment when Amazon is also defending its service reputation against faster-growing rivals.
Selling partner services and the seller experience
Selling partner services is the function third-party merchants deal with when an account is suspended, a listing is suppressed or a reimbursement is disputed. Sellers pay referral fees, fulfilment fees and advertising fees, and seller support is the service they receive in return.
Cutting that team while marketplace fee revenue grows is the part of this story most likely to generate friction. Seller advocacy groups have complained for years about response times and appeal quality, and a smaller team invites the obvious inference.
Amazon’s counter-argument would be that most seller contacts are routine and automatable, and that the remaining cases need better tooling rather than more headcount. That is a testable claim, and the test runs through the holiday quarter.
The parts of Stores that are still growing
Not every corner of Stores is shrinking. Amazon has been staffing up a separate urban fulfilment and dark store effort, and our reporting on the Amazon Now dark store build-out found net-new design and pre-opening roles rather than cuts.
That is the usual shape of an Amazon reorganisation: support and coordination layers contract while a named growth bet absorbs the budget. The company’s language about “structure” and “priorities” is doing a lot of work in that sentence.
How do Amazon’s cuts compare with the rest of retail in 2026?
Amazon is not an outlier this year. Corporate and operational reductions have run across US retail and logistics through 2026, with most companies citing structure, automation or a reallocation of spend toward stores and technology.
| Company | Reported action in 2026 | Roles | Where the cut landed | Stated rationale |
|---|---|---|---|---|
| Amazon | October 7 to 8 | Fewer than 1,000 | Stores division; US, UK, India | Structure better suited to priorities |
| Walmart | May 12 (largest round) | About 1,000, roughly 1,100 for the year | Global Tech, e-commerce fulfilment, Walmart Connect | Structural review; relocation to Bentonville or Northern California offered |
| Target | Confirmed during 2026 | 100 plus 400 | District offices and supply chain | Reinvest in store labour hours under a new chief executive |
| Starbucks | Corporate round in 2026 | 300 | US corporate and regional support offices | Organisational turnaround; store staff unaffected |
| UPS | Plan stated for 2026 | 30,000 | Operational workforce, largely full-time drivers | Network reconfiguration, mostly attrition and voluntary separation |
Two patterns stand out. First, the retail cuts are concentrated in corporate, technology and supply chain coordination roles, not in stores. Target was explicit about the trade: fewer district office and supply chain roles, more hours on the shop floor.
Second, the logistics cuts are an order of magnitude larger and operational in nature. UPS’s stated plan to remove 30,000 operational roles in 2026, largely through attrition and voluntary programmes for full-time drivers, is a network decision rather than a head-office one.
Against that backdrop, Amazon’s sub-1,000 action is small. What makes it notable is the division it lands in and the week it lands in, not its size.
The common thread across all five is that none of these companies described the cuts as a response to weak demand. Every stated rationale is about structure, automation or where the money should go instead. That is a different message from a downturn layoff, and it should be read differently.
Third-party trackers put cumulative US retail job cuts for 2026 in the ten-thousand range, though those aggregates mix corporate reductions, store closures and logistics restructuring, and they are not comparable with company disclosures. Treat them as directional rather than precise.
What does this mean for marketplace sellers and brands?
The practical exposure for sellers is support latency during the busiest trading weeks of the year. If seller support headcount is lower and contact volume follows its normal seasonal curve, queue times are the variable that absorbs the difference.
Three categories of case carry real money. Account health actions that take a seller offline, listing suppressions on a hero ASIN during a deal event, and inventory reimbursement disputes after a fulfilment error. Each is time-sensitive in a way that a routine query is not.
The sensible response is procedural rather than dramatic. Open cases earlier than you otherwise would, document inventory discrepancies as they occur rather than in a January reconciliation, and keep a second sales channel live so a suspension is a revenue dent rather than a revenue stop.
Channel diversification is also where the platform layer is moving. Amazon’s decision to block Meta’s Muse AI agent from its storefront while Shopify opened Shop Pay showed how quickly the routes into a catalogue can change. Sellers who depend on one support queue and one demand source carry both risks at once.
The fee load versus the service load
The uncomfortable arithmetic for sellers is that the fee schedule has moved in one direction while support staffing has moved in the other. Referral, fulfilment, storage and advertising charges have all been revised upward or extended over recent years.
Amazon would argue those fees price the fulfilment network and the demand it provides, not the support queue. That is a defensible position, and it is also why seller support is structurally vulnerable: it is the one part of the bundle that no fee line explicitly funds.
What changes for brands running deal events
Brands running promotions through Amazon’s event calendar carry a narrower version of the same risk. A deal that goes live with a suppressed variant, a broken parent-child relationship or a coupon that fails validation needs resolution in hours, not days.
The mitigation is unglamorous: lock creative and listing changes well before an event window opens, verify every variant renders on the live detail page, and avoid making structural catalogue edits in the 72 hours before a deal goes live.
What should you watch at the October 29 earnings call?
Amazon reports third-quarter 2026 results on Thursday, October 29, with the conference call at 5:00pm Eastern. The company guided to net sales of $197bn to $202bn for the quarter, equivalent to year-on-year growth of 9% to 12%. Analyst consensus has sat near the top of that range.
| Marker | Figure | Why it matters here |
|---|---|---|
| Q3 2026 net sales guidance | $197bn to $202bn | Sets the bar the smaller Stores organisation is expected to clear |
| Consensus revenue | About $202bn | Top of guidance; little room for a service-driven miss |
| Implied growth | 9% to 12% year on year | Volume is rising while support headcount falls |
| 2026 capital spending | About $220bn | Shows where the budget is going instead of headcount |
| Total employees, Q2 2026 | About 1,595,000, up about 3% year on year | Corporate cuts of this size are invisible at group level |
Three things on the call would confirm the structural reading. A question answered with specifics about customer contact deflection rates. Any disclosure of a corporate headcount figure separate from the operations total. And fourth-quarter guidance that assumes a lower cost-to-serve rather than a higher one.
One thing would undercut it. If Amazon’s fourth-quarter outlook includes elevated service or fulfilment costs, the October cuts look less like a confident automation call and more like a budget action taken under pressure.
It is also worth watching what Amazon does not say. The company has never broken out corporate headcount from its group employee total, and there is no obligation to start. If the October action is the template, the only durable record of corporate shrinkage will be journalism plus the gap between revenue growth and the group total.
That gap is already visible. Revenue growing at 9% to 12% year on year against group headcount growing at about 3% means revenue per employee is rising, which is the metric the restructuring is designed to move. The question is whether it rises because the work got more efficient or because less of it is being done.
How do headcount cuts and a $220bn capital plan fit together?
The apparent contradiction resolves once you separate the two budgets. Amazon’s capital spending plan of roughly $220bn for 2026 is concentrated in data centres, chips and fulfilment automation. Corporate salaries sit in operating expense, and that is the line being trimmed.
Group headcount tells the same story from the other side. Amazon reported about 1,595,000 full-time and part-time employees as of the second quarter of 2026, up roughly 3% year on year, even after 30,000 announced corporate cuts. The operations workforce grew faster than the corporate workforce shrank.
That composition shift has been visible in pay policy too. Amazon’s move to raise minimum pay to $20 an hour with a grocery discount, a commitment of about $1.5bn, was aimed squarely at the hourly operations base rather than at salaried staff.
Read together, the picture is a company spending more per operations employee and on physical and compute capacity, while holding or reducing the salaried layer that coordinates it. Whether that is efficiency or a bet that has yet to be tested depends on the fourth quarter.
One further asymmetry deserves attention. Automation savings land in the quarter the headcount leaves, while the service cost of under-resourcing lands one or two quarters later, in returns, concessions, refunds and churn. A reduction can therefore look successful on the first reporting date and expensive on the third.
That lag is why the fourth quarter is the real test rather than the third. Contacts generated by December delivery exceptions are resolved in January, and the associated concessions show up against the quarter after that.
What does this story not tell us?
It does not establish a number. Amazon has not confirmed a total, and “fewer than 1,000” is a boundary rather than a count. Anyone citing a precise figure is citing a reporter’s estimate.
It does not establish a cause. Automation, cost discipline, a strategy change inside Stores and a simple duplication of roles after two large reorganisations are all consistent with what has been reported, and the company’s statement distinguishes between none of them.
It does not establish a trend. One sub-1,000 action nine months after a 16,000-role tranche is a single data point. A second comparable action before year-end would make it a pattern; absent that, it is a reorganisation.
Finally, it says nothing about the holiday peak workforce. Seasonal operations hiring runs on a separate budget and a separate calendar, and nothing in this reporting touches it.
FAQ
How many jobs did Amazon cut in October 2026?
Fewer than 1,000 corporate roles, according to reporting from Reuters and Business Insider. The Information described hundreds of layoffs in the Stores division specifically. Amazon has not confirmed a figure, so the total should be treated as a reported estimate rather than a disclosed number.
Which Amazon division was affected?
The Stores division, which runs Amazon’s retail and e-commerce business. Internal posts described to reporters named customer service and selling partner services among the teams affected, with smaller reductions elsewhere in the company.
Which countries are affected?
The United States, the United Kingdom and India all appear in the reporting. Indian outlets led with the local angle, noting that staff in Amazon’s Indian retail organisation are among those losing roles.
Does this affect warehouse or delivery jobs?
No. Everything reported concerns corporate, salaried roles. Fulfilment centre and delivery station staffing runs on a separate budget and normally expands ahead of the holiday peak, and nothing in the reporting touches seasonal operations hiring.
How does this compare with Amazon’s earlier layoffs?
It is much smaller. Amazon announced about 14,000 corporate cuts in October 2025 and about 16,000 in January 2026, roughly 30,000 in total and the largest workforce reduction in its history. This action is under 1,000, around 3% of the January tranche.
Did Amazon break its promise not to keep cutting?
Not on a literal reading. In January, Beth Galetti said Amazon was not trying to create “a new rhythm” of broad layoffs every few months and that “that’s not our plan.” A sub-1,000 divisional action nine months later is neither broad nor every few months. She did, however, say teams would keep evaluating and making adjustments, which describes this kind of event.
Why did Amazon cut jobs during Prime Big Deal Days?
Amazon has not explained the timing. The cuts were reported as the October 6 to 7 event closed and three weeks before third-quarter results on October 29. Either the calendar collided with a planned fourth-quarter reorganisation, or the company waited for the event to finish before moving ahead of earnings.
What should Amazon marketplace sellers do now?
Assume slower seller support during the holiday quarter and plan around it. Open account health, listing suppression and reimbursement cases earlier than usual, document fulfilment discrepancies as they happen rather than in a later reconciliation, and keep a second sales channel live so a suspension costs you margin rather than all of your revenue.
When will we know whether the cuts hurt service?
The fourth quarter is the test, and the October 29 earnings call is the first checkpoint. Watch for any detail on customer contact deflection and for fourth-quarter guidance that assumes a lower rather than higher cost to serve. Elevated service or fulfilment cost assumptions would suggest the reductions were a budget decision rather than an automation one.