It’s a common misconception that testing brings the best return on investment when it’s confined to the latest and greatest device and browser versions.
For example, conventional wisdom would suggest that as the Google Chrome browser updates itself automatically, every user is bound to be using the latest version, and this is where testing should be targeted. In reality, Digivante’s analysis of over a hundred companies shows that this is not the case, for a variety of reasons, including users turning off automatic updates because they want greater control.
Device-browser proliferation means testing on multiple versions of every browser – not just the most recent – has become a matter of urgency for website and app developers.
In the same way, you cannot assume that smartphone users always upgrade to the latest release. Many now fear that the latest operating system updates will slow their devices and some older, but still popular, devices are limited to the only operating system version they can run. This means that there are likely to be more devices with non-current operating systems than you’d expect.
For example, although iOS 14 became available to download in September, you may have owners of the iPhone 6, with its limited processing power, using your website or app. After all, they remain the largest group of iPhone users – not that you would be able to monitor this with Google Analytics tracking.
What might seem an arcane topic hit the headlines recently, when people attempting to download the NHS coronavirus contact tracing app onto older iPhone’s found that the app requires iOS 13.5 or later.
“iPhone 6 users furious as NHS track and trace Covid-19 app does NOT work with their phone”
Daily Express, 24th September 2020
Are you bridging the digital divide?
Being a user of an older device says little about the owner’s spending power. So you may be missing out significantly on potential business, without even knowing it.
Demographics play a part in all this. Youngsters may inherit their parents’ cast-off devices until they become economically active in their own right – but they still have pester power in spades! Adult children may pass their older technology on to parents who don’t need lots of bells and whistles, just a smartphone that works.
Professionals in the middle age range are most likely to have an up-to-the-minute device. But even if this is your core market, you still cannot guarantee that all your prospects and customers are using the latest technology. You de-prioritise testing on older devices and browsers at your peril. Certainly, any company targeting the mass market needs to branch out and test its app or website on a range of devices, to ensure that everyone can use it to a satisfactory level.
Users in the older age bracket are, incidentally, a growing ecommerce target, as they switch to online shopping in droves to avoid the Covid-risky high street. This creates a fantastic opportunity to turn a crisis into an opportunity and is not to be overlooked.
Testing for the real world
Clearly, spreading your net more widely in testing is a wise policy. However, the only way to obtain valid results is to test on actual older devices or browsers. Emulation tools such as BrowserStack cannot reflect the real user experience.
Here at Digivante, we don’t use emulation or simulation. We test manually. We give our global community of testers the assignment of testing on their older devices and using older browsers; then we reproduce the issue onto a newer device.
The client’s development team can then assess the likely impact and set their priorities: is the issue confined to a very small subset or does it have wider implications? Testing on older devices and browsers may also bring to light issues to which newer technology is not immune.
Having a global community of testers means we can deliver at warp speed. It took our testers just 72 hours of exploratory testing to uncover 67 issues (8 around conversion and 59 affecting the customer experience) for Figleaves, a UK-based online lingerie and swimwear retailer. We tested 708 variations of device, operating system and browser combinations.