In my last article, Are You Ready for Growing Mobile Workloads, I asked you to keep an eye out for Black Friday and Cyber Monday success and failure this year. I was really curious to see if retailers had learned their lesson from past years’ online retail disasters. Well, for this year, the answer seems to be yes and no. Not surprisingly, there were a few retailers who experienced failure for the first time (or for the first time that we noticed…) and some retailers with past bad experiences learned from their mistakes and escaped this year’s Black Friday and Cyber Monday blitz unscathed. Incredulously however, there were others that had repeated failures and clearly had not planned properly.
On Black Friday, we learned that Neiman Marcus, Newegg, HP and Jet.com all had online failures to some extent, according to InternetRetailer.com. There was a fury of activity, over 36% of which resulted from mobile shopping sources. But Black Friday didn’t seem as disastrous as events from past years, as retailers made plans to make significant gains on Cyber Monday, with promotions abounding.
Cyber Monday was a different story, however. It seems that Target once again had trouble keeping up with the demand, a repeat of their past troubles. But it wasn’t just them – Neiman Marcus, Best Buy, Walmart, and eBay all suffered from some level of outage on the most important retail weekend of the year.
So what is the problem? Part of it is just not being prepared with enough IT computing capacity to handle the expected spike in transaction processing –which is really inexcusable, considering the importance of the business, and past retail history managing that business. Part of it can also be attributed to criminal activity – hacking, denial of service attacks, and the like. But it’s not like DoS is a new threat; if a retailer is hit by this type of failure, then they really need to look at their own lack of preparedness. A new, perhaps unexpected problem could be an upsurge in purchase transactions from China.
The thing is that some of these problems could be addressed nicely using computing platforms that make increasing computing capacity easier to manage. Proper planning is a definite requirement no matter what type of platform you’re running, but transaction processing systems that run on the mainframe platform have a built-in advantage – they can be configured to increase system capacity very quickly – in some cases automatically. But platforms aside, better planning could have helped these retailers get through the most important retail weekend of the year without running into computing capacity issues—which is another way of saying without leaving money on the table. Heads should roll…