Macy’s wasn’t the only retail website that was overwhelmed on the biggest shopping day of the year. Old Navy, GAME, and Quidco were also caught unprepared for the high traffic surges, and lost sales opportunities in the short and long run.
In 2015, it was Neiman Marcus and Target that collapsed under the weight of online demand, and the year before, Best Buy.
The holiday season will soon be upon us, and it is important for e-commerce companies to assess their readiness for the onslaught of online shoppers. IT infrastructures that currently run smoothly supporting normal day-to-day traffic may not be prepared to handle the surge of holiday shoppers. Can your applications and website support this increase in activity?
Lesson’s from Amazon
Let’s look at how Amazon itself prepared for its own Black Friday –Amazon Prime Day – when the web giant gives special deals to its most loyal customers. According to the company, Amazon accounted for 74 percent of all U.S. consumer e-commerce on Prime Day 2016 and saw record-high levels of traffic including double the number of orders on the Amazon Mobile App compared to Prime Day 2015.
Scaling EC2
Amazon prepared for the traffic spike by employing a small army of EC2 instances to handle web traffic, using resources from its availability zones around the world. It then prepared for surges from time zones throughout the world as people started their shopping from Asia to Europe to the United States.
Amazon also meticulously prepared and planned for the event. The retail IT team looked at historical performance to predict traffic, and staged failure exercises by intentionally (and unexpectedly) breaking parts of the infrastructure to run drills. It also automated as much scalability as possible, using Route 53 to automatically scale DNS, Auto Scaling to scale EC2 capacity on-demand, and Elastic Load Balancing to automate failover and to automatically balance traffic in every region. Finally, a dedicated team monitored CloudWatch with religious fervor to track usage and ensure customer satisfaction.
Expecting the Unexpected
Unfortunately, Black Friday-like experiences can happen to a company unexpectedly. For example, prior to the release of Star Wars: The Force Awakens, we helped MovieTickets.com migrate its entire customer-facing infrastructure to Amazon Web Services and tested it repeatedly. Fandango, another online ticketing company, was not so well-prepared. Its site crashed when tickets went on sale, and shoppers were placed a purgatory-type waiting room, where they had to wait all night if they want to secure tickets.
Learning from Mistakes
Learning from the mistakes and successes of other companies when demand for web resources exceeds expectations can help you prepare for planned and unplanned events. Black Friday, Cyber Monday, elections, global media releases, product launches, and other events can wreak havoc on you and your customers. They can also give you an opportunity to delight them with a positive experience while others are frustrated on competing e-commerce sites .
Onica is a top AWS Premier Consulting and audited Managed Service Partner. We’ve migrated 85,000+ servers to AWS, performing everything from basic “lift and shift” to helping to re-architect and manage services on clients’ infrastructures so that they can take full advantage of cloud resources. If you’d like to learn how working with an AWS Premier Consulting Partner can improve your business, contact us for a quick assessment.
