Amazon’s website and shopping app went down for thousands of users on Thursday afternoon after a faulty software code deployment knocked out checkout, login, and product pricing across the platform. The disruption lasted roughly six hours before Amazon confirmed the issue was resolved.
Amazon down reports surged on March 5, 2026, starting just before 2 p.m. ET and rapidly escalating into one of the platform’s most widespread public outages in recent memory. Reports spiked from just 316 issues at 1:48 p.m. ET to 10,943 at 2:03 p.m., reaching a peak of 21,716 reports at approximately 3:48 p.m. The Shortcut The issues affected both the website and mobile apps simultaneously, leaving shoppers unable to complete purchases and sellers staring at inventory pages with no images and no listings.
Amazon confirmed the incident was due to “a software code deployment.” Spokesperson Jennie Bryant stated: “We’re sorry that some customers may have temporarily experienced issues while shopping. We have resolved the issue, which was related to a software code deployment, and website and app are now running smoothly.” The issues appeared to be largely resolved by 8 p.m. ET. Startup News
What the Amazon Down Reports Actually Showed
Downdetector data showed 58% of problems related to checkout, 17% to product pages, and 14% to the mobile app. Khaleej Times The Washington Times reported a slightly different breakdown from later in the outage window: nearly half of reports, 48%, concerned the checkout process, with 21% relating to the mobile app and 17% to individual product pages. The Shortcut
While Amazon’s homepage remained accessible, prices and product information failed to load, and clicking through to product pages in many cases returned an error page reading: “Sorry, something went wrong on our end.” 91Mobiles Many users encountered Amazon’s signature dog error pages — the company’s standard fallback when pages crash — spread across multiple sections of the site.
Users reported checkout and payment errors, incorrect or fluctuating prices, and problems updating shipping locations. Khaleej Times Amazon Fresh, the company’s grocery delivery service, was also affected, with users reporting issues placing orders or viewing their purchase history. Startup News
Login failures added another layer of disruption. Some users said they could not log in through any device, that product pages slowly lost their ratings and reviews before displaying nothing, and that the platform logged them out without allowing them back in. Smartprix
Sellers Were Hit Just as Hard
The disruption was not limited to shoppers. Third-party sellers using Amazon’s Seller Central platform reported significant inventory problems.
Sellers described scrolling through their inventory pages only to find 50% of product images reading “image not available” and approximately 75% of total inventory going offline. Clicking on affected product pages returned the same error messages shoppers were seeing on the consumer side. Smartprix
Amazon’s Help account on X acknowledged the issue with a standardized message: “Hello. We’re sorry that some customers may be experiencing issues. We appreciate your patience as we work to resolve the issue.” Tech Advisor Beyond that, Amazon provided no real-time explanation of what had gone wrong until the post-incident spokesperson statement confirmed the software deployment as the root cause.
For sellers who rely on continuous platform visibility to generate sales, even a few hours of downtime can translate into measurable lost revenue — particularly for those running active advertising campaigns tied to live product listings.
What Caused the Amazon Down Incident
The draft theories about cloud infrastructure failures, microservice cascades, and authentication database problems can be set aside. Amazon’s official explanation was precise: a software code deployment went wrong.
The company’s website experiencing issues without a larger AWS outage is somewhat unusual, and suggests the problem originated outside of its cloud infrastructure rather than in the underlying server architecture. 91Mobiles AWS powers a significant portion of the modern internet, and a true cloud-layer failure typically produces a much wider blast radius across unrelated services.
This outage appears to have been a deployment error — the kind of incident that can happen when code pushed to a live production environment introduces an unexpected bug that breaks dependent systems. In Amazon’s case, the pricing and checkout systems appear to have been the first to surface problems, followed by login authentication and product page rendering.
Amazon has experienced its fair share of outages as a cloud provider through AWS, including one in October 2025 that took out services like Snapchat and Amazon’s own Alexa voice assistant for hours. 91Mobiles Thursday’s incident was a different animal — localized to Amazon’s retail platform rather than its cloud infrastructure layer, and ultimately resolved within the same business day.
What Happens After an Amazon Down Event
Amazon did not release a public post-incident review, which is standard practice for retail outages compared with AWS events, where detailed service health reports are typically published. The spokesperson statement confirmed the software deployment cause and declared the platform restored.
For third-party sellers, the absence of communication during the outage is a recurring frustration. Seller Central forums filled with complaints about inventory disappearing and order pipelines going silent with no timeline from Amazon about when service would return.
For consumers, the practical impact on any individual transaction was limited — most shoppers who encountered errors could return later that evening to complete their purchases after the 8 p.m. ET resolution. The broader significance is what the scale of reporting says about dependency: more than 21,000 Downdetector reports within two hours illustrates how quickly a platform used by millions registers even a partial disruption.
Amazon has not announced any changes to its deployment practices as a result of the incident.
FAQ
Q1: Why was Amazon down on March 5, 2026? Amazon confirmed the outage was caused by a software code deployment error. Spokesperson Jennie Bryant stated the issue was “related to a software code deployment” and confirmed the platform was restored by approximately 8 p.m. ET.
Q2: How many users reported the Amazon down outage? According to Downdetector, reports peaked at 21,716 around 3:48 p.m. ET on March 5, 2026. The spike began just before 2 p.m. ET, climbing from 316 reports at 1:48 p.m. to nearly 11,000 within fifteen minutes.
Q3: What problems did users experience during the Amazon outage? Users reported missing product prices, checkout and payment failures, login errors, problems updating shipping information, and issues with Amazon Fresh grocery delivery. Sellers using Seller Central reported inventory pages going offline and product images disappearing.
Q4: Was AWS responsible for the Amazon down incident? No. Amazon’s web services cloud infrastructure was not implicated in this outage. The disruption was isolated to Amazon’s retail platform and traced to a software code deployment, not an underlying cloud infrastructure failure.
Q5: How long did the Amazon outage last? Reports began surfacing around 1:55 p.m. ET and the platform was largely restored by approximately 8 p.m. ET on March 5, 2026 — a disruption window of roughly six hours for the most affected users.
