Don’t Be the Next United, NYSE or WSJ: The Case for Combating Software Failure with Software


The above image is taken from the movie The Truman Show. While I’m not a fan of the film (give me Edtv over Truman Show any day of the week!), I found the above clip relatable.

See, Truman here is at a travel agency, and this travel agency has a blatant agenda against travel because they want to discourage Truman from leaving his manipulated life-set (No spoiler alerts for a 17-year old movie—if you haven’t seen it yet, that’s on you).

I’m here trying to point to software failure as a reason to buy more software. My intentions are a lot less manipulative than the producers of The Truman Show, I swear.

Let’s look at United Airlines. We’ve all dealt with the customer-facing technology utilized by United and similar airlines. While on-board entertainment screens can be a little frustrating at times, for the most part, web portals and booking interactions are as up-to-date as you would expect from large enterprises (again, more or less).

Behind all that customer-facing sheen, however, sits a convoluted mess of legacy systems and mainframes tied together by patchwork coding. There is no single source of truth, and it takes almost all of their developers’ time to just keep the beast alive. Given the complex intricacies of these infrastructures, it’s no surprise when things occasionally go down.

These types of entangled predicament is the child of short-term planning and win-now strategies, which, like it or not, is the cold reality facing just about every business that has ever existed.

Luckily, there’s some smart people that have developed solutions against legacy system predicaments like United’s. Legacy systems aren’t going anywhere—the cost of replacing systems like-for-like is just too debilitating—but there are smarter ways to clean and normalize data from legacy sources to prevent the kind of confusion that causes operations to shut down.

The next-generation of process-oriented applications, like Colony Logic’s Swarm CPQ and Harvest Analytics, sit on top of architectures, controlling data and making it work for your sales/operations/reporting needs. Your legacy systems don’t need to be ripped out—they can simply serve the one function they will always do well: housing and sorting data.

July 8th will be a black mark for United, NYSE and WSJ for a little while. Circling back to Truman, it could happen to you too!