Software Testing

Pendo trawls software activity for usability clues

‘User-friendly’ is a term that has fallen out of favour to the extent that it now seems a relic of the 1980s. But that’s not because all software became easy to use: very often it’s quite the reverse as the rise of new platforms and operating systems has led to a smorgasbord of different user experience approaches.

This is a situation that Pendo, a three-year-old startup headquartered in Raleigh, North Carolina, is trying to improve by helping software vendors figure out the issues users have with their programs.

Pendo monitors how users interact with software via a subscription-based cloud service. The mission, says Pendo CEO and founder Todd Olson is to “make software experiences better”. Olson, an experienced operator formerly of Borland and several other tech companies, wants to help unlock the underlying richness and features that too often remain undiscovered or indecipherable to users.

Although Pendo is cloud-based some vendors use the service for on-premises software. Whatever the platform, the idea is that they can address usability issues by changing features, adding help such as ToolTips or walkthroughs, exposing features more clearly, surveying users and so on. But after decades of usability testing why, I ask, is Pendo necessary?

The answer lies in the benefits of ongoing testing rather than the old upfront, pre-release usability testing model. “I’m not here to say what we do replaces usability testers,” Olson says. “What’s different with cloud-based software is we have the ability to test, but day in day out, the ways people use software. We augment, not replace [orthodox] usability testing.”


Democratising user testing

Pendo also brings the benefits of a data-centric, independent view of how software is being used for companies that can’t afford legions of user experience experts

“Companies like Google, Facebook and Microsoft have been doing it for years but most companies are not Google, Facebook and Microsoft,” Olson adds.

Software is an interesting case study in the context of IT history: while Moore’s law has seen compute performance increase, we have often gone backwards and forwards in the usability of software. There have been various attempts to address this. In the 1980s IBM tried to create a blueprint for how software looks and feels, called Common User Access (CUA). Microsoft’s Office Compatible Program in 1994 was an attempt to make all Windows applications look like Microsoft applications. But today the proliferation of devices, input types, screen sizes, system software and cloud services, together with a rapid software release cadence and community development, means that the “look and feel” of software has become an anything goes, Wild West affair. But for Olson velocity and change are not negatives.

“The pace of innovation is so much faster and you can iterate much quicker,” he says. “You can put something out there, collect feedback, take polls and iterate on the same day.”

The more predictable way of asking people what they want to see in software isn’t even necessarily a good guide, he believes, and even idea management software such as Dell’s IdeaStorm and Salesforce’s IdeaExchange cannot provide the full picture. As for the old release schedules of the boxed software epoch, they were effectively asking users to predict what they might want nine months down the line.

“There’s what people ask for and then there’s what people actually use,” Olson says. “When you ask them what they want, you’re asking them to predict the future.”


Rights and wrongs

Pendo’s reach into the behaviour of about 15 million users running software from the likes of Infor, BMC, Coupa and Citrix gives it a good overview of how people use software today. So, I ask Olson: what are companies doing wrong when they design software today?

The answers are application-dependent, he says, but he does seem to have a few bugbears. One of them is using visuals when words might be a better clue: think of Apple’s ‘hamburger’ icon on mobile apps instead of the word ‘Menu’, for example. “At the end of the day, it’s a fact that more people can read than guess the meaning,” he says.

Another annoying wrinkle is the assumption every user will use the same functions in the same way and yet another is the idea of adding more ways of interacting with the screen via taps, double taps, presses and so on. “A lot of people will strive for the minimalist approach but what I don’t like is if you drive for that over discoverability… I don’t like to guess on apps that have no help, no guidance and you need to know you have to make a double finger swipe.”

Speaking of minimalism, UI designs have had more new trends than Paris catwalks and we’ve seen our screens switch from skeuomorphic to minimalist in the last few years. One challenge is that ‘lite’ versions of programs never seem to sell well despite the old chestnut that we use only a small minority of the software pay for. But Olson spies a halfway house that cuts clutter without sacrificing features.

“Taking things away provides some of the best data you get,” Olson enthuses, but he’s also a fan of “progressive disclosure”: front-ends that only reveal certain features on certain cues rather than bombarding users with the full torrent of features from the off.

Companies like Slack, he says, have done a fine job of creating a “crawl-walk- run” familiarity sequence using this tactic. Pointed there by his kids he also likes Snapchat (“pretty neat, very clean, very smart”) but he also admires InVision’s designer tools, calls the Medium publishing platform “amazingly beautiful”, and adds that he’s a Microsoft and Google UX fan.

Olson plays a straight bat to my suggestion that some of Apple’s software designs are bewildering; he uses an iPhone and a Mac. But Yahoo, he says when I raise its universally derided Calendar “upgrade”, has “lost a lot of its strength”.

There’s scant likelihood of the interfaces we use coalescing to form an orderly and coherent face. In fact, with wearables, VR, AR and the Internet of Things we’ll probably need companies like Pendo more than ever. Pendo’s far from alone in its space but its product focus and insistence on being data-driven might give it more focus than some rivals. With over $33m collected in VC it has a chance of being the engine under the hood of the next generation of software.


Also read:
Design guru Don Norman slams Apple’s ease-of-use ‘disservice’
Don Norman: Microsoft is beating Apple at design


« Typical 24: Julie Baxter, CA Technologies


JustElect is forging a path for electronic voting in Ghanaian institutions and beyond »
Martin Veitch

Martin Veitch is Contributing Editor for IDG Connect

  • twt
  • twt
  • Mail

Recommended for You

Trump hits partial pause on Huawei ban, but 5G concerns persist

Phil Muncaster reports on China and beyond

FinancialForce profits from PSA investment

Martin Veitch's inside track on today’s tech trends

Future-proofing the Middle East

Keri Allan looks at the latest trends and technologies


Do you think your smartphone is making you a workaholic?