I am still amazed how so many high profile sites go down or experience performance degradation year on year due to inability to handle a sudden flood of traffic. What’s even more disturbing is some sites may stay online while experiencing significant performance degradation, this issue often doesn’t trigger any alarms and in some cases is not monitored by organisations. However, the evidence shows many customers feel performance degradation is as good as a complete outage.

I recently came across a study by Dynatrace which highlighted “75% of all smartphone/ tablet users and 81% of Millennial smartphone/tablet users report that if a mobile site or mobile app is buggy, slow or prone to crashes, they would abandon it and shop elsewhere“. This survey was completed slightly over a year ago, so the numbers have probably increased further.

They study also showed Australian Millennials are the most likely to vent their frustrations publicly, more than any other country surveyed. Over 59% of Millennials surveyed said they are likely to complain via social media about a poor online shopping experience. It was not just the Millennials either, Australia was in second place for Adults (18+) with 46% saying they would vent on social media.

Moreover, what about the costs? Standish Group calculated, the average cost per minute of downtime for a point of sale website to equal USD $4,700 per minute (AUD $6128)!

Many people think Performance Testing is just a check to see if you can survive a flood of traffic; However, there’s much more to it. Comprehensive Performance Testing will not only identify if you can survive a flood of traffic and to what point but also highlights the quality of service you are delivering as load increases, in addition to service quality over longer durations. You gain deep insights into your system behaviour, learn about weaknesses before they are reported by your customers on social media and have the rare opportunity to do something about it in advance.

As I highlighted earlier, customers do not care if your service is still online, if it is running slowly many will abandon. Conversely, If your site, app or system already scales well – how do you know you have not overdone it without testing? We have seen situations where organisations are paying >$100k in cloud hosting services but only running at sub 5% load during peak periods, in a scenario like this an appropriately crafted testing program can help significantly reduce infrastructure costs while maintaining a suitable level of contingency.

Thankfully, an increasing number of Australian organisations are starting to take this seriously by implementing Performance and Load Testing into their processes, and by enabling real-time Application Performance Management.