Tallying the damage: This week's AWS outage hit retail hard

Amazon (flickr)
Amazon has acknowledged that the outage was due to human error.

When Amazon Web Services experienced an outage earlier this week it wasn’t just Netflix and Reddit that were affected. Online retailers were among those most negatively impacted when AWS’ cloud storage service, S3, went down on February 28.

So, what was the final hit to retailers? It’s hard to know for sure, but Cyence, a startup focused on measuring cyber risk, estimates that S&P 500 companies lost $150 million from the four-hour outage.

Business Insider reports that three major retailers went down completely—Express, Lululemon and One Kings Lane—while slow load times proved costly for others.

Web-monitoring company Apica told Business Insider that more than half of the top internet retailers saw performance decrease 20% or more. The website load time for Disney Store, for example, took 1,165% longer than typical and Target took 991% as long.

Amazon’s role as a near-ubiquitous hosting provider meant that the outage crippled a sizable chunk of the internet—this website included—and may be a wake-up call for retailers that find themselves overly reliant on a single vendor.

In a postmortem on the event, Amazon acknowledged that the outage was due to human error when an employee trying to debug an issue with the intention of improving the performance of S3’s billing system took several servers offline, which had a cascading effect on a number of subsystems and virtual machines.