This was originally posted on LinkedIn on November 24th 2016
My wife forwarded me a news story yesterday about a possible performance testing lead. Adele is coming to New Zealand to perform three concerts, and the tickets sold out in less than two hours. The huge burst of interest brought Ticketmaster’s website to a crawl and fans were outraged at being unable to secure a ticket.
Without a doubt, this is an example of production performance issue. But get ready for a controversial opinion: in my view, there is no business case to invest any time to resolve it.
The bottom line is that (most) performance testing is about protecting revenue. Did the slow-down impact Ticketmaster’s sales? No. Were there any competitors also selling tickets? No. As long as the Ticketmaster website is capable of selling all available tickets within a reasonable time-frame, there is no financial incentive for them to improve website performance.
In the long run, it’s not good for Ticketmaster’s brand reputation. However, given the nature of their business model (having a ‘monopoly’ of the sales of each event they handle sales for) it probably won’t do them any financial harm.
To be fair, this is a unique example. Far more often businesses underestimate the impact of performance issues and how they could hit their revenue and reputation, and they incur substantial losses as a result.
Especially in New Zealand (or any smaller economy), there is huge value in understanding the business risk before you start performance testing. It helps you understand how much performance testing is appropriate.