Perfecto Mobile, the Boston-based provider of testing services for mobile app developers, has raised $35 million that it will use to expand its products and build out its sales staff. The investment comes from Technology Crossover Ventures along with existing Perfecto investors FTV Capital, Carmel Ventures, Globespan Capital Partners and Vertex Ventures.
In particular, Perfecto said it will grow its test lab capability from running native application tests on cellphones and tablets to include web application testing. Perfecto runs more than 1.2 million tests monthly on mobile devices in its lab, it added, which allows customers to test their applications via the cloud on real devices.
Amir Rozenburg, Perfecto
Developments in mobile technology are creating a “staggering, explosive” growth opportunity in mobile trading, according to Amir Rozenburg, product manager and evangelist at Perfecto. Wearable devices like the iWatch are also going to encourage more people to trade remotely, said Rozenburg, if only because investors stuck in a business meeting at work will be able to use their watch discreetly to trade when an alert that buzzes through. It’s simply another way that investors really can be connected to their portfolio on a second-by-second basis, said Rozenburg.
But what happens if you are trying to trade during your daily commute on the train, and your connection switches from 4G down to 3G? If a switch to 3G means it takes nine seconds to log in to a trading platform, then a good investment idea may well have become loss-making. Perfecto’s business in testing apps was born of the mis-match between what testers were doing in the lab and the conditions experienced by the user.
The core testing methodology revolves around the personas created to replicate those users. To adjust a test, a line of code is added according to the chosen persona. The beauty of the methodology is that Perfecto has found that a limited number of personas can cover the vast majority of users, and that a typical app has a target audience that can be represented by two or maybe three personas.
There’s “Georgia” for example: a travelling salesperson who is based in New York City, whose mobile use is frequently interrupted by incoming calls. There’s “Peter”, a municipal worker who lives in the UK, uses an older mobile phone and experiences regular network disconnects. “Sarah” is a stay-at-home mother who does a lot of social networking, and “Sam” is a twenty-something gamer.
As for the retail stock investors targeted by mobile app developers, the specific tests can be covered by two principal personas, according to Rozenburg. One is more elderly, moves around less, and uses the mobile device mostly for researching stocks. The other is younger, moves around more, and trades more.
Whatever the accuracy of the persona tests, the challenge of unreliable network connections is here to stay for a while, said Rozenburg. So developments in the archiving and storage of data that can be used to reconstruct interrupted trades, along with authentication processes that ensure traders are aware that they are resuming the interrupted trade, are going to be a focus for developers.
