Tuesday, March 26, 2013

Testing, Testing, 1-2-3

Alas, there comes a time in the lifecycle of all software, when the project must be tested. Keeping the test in mind while building a project can help a programmer ward off potential pitfalls throughout the process, so it is important to begin thinking about testing early, even during the design process.

Since Martin and I have decided to build a web app, we will have the option to utilize two types of testing. The first method will involve inviting friends to test the app on our own individual devices (that is, Martin's phone and my phone). This will act as a pilot study, pointing out any major flaws in usability.

Once we have finished the pilot study, we will progress to larger-scale testing. This will be done by soliciting feedback on a shared version of our app that can be run on a mobile emulator.

We will look at the following things when we conduct testing:
 -Does the app fail?
 -How long does the average user spend in the app?
 -In which areas of the app does the user spend the most time?

In addition to the above objective questions, which will be measured behind-the-scenes, we will solicit the following subjective material, to be rated on a scale to be defined for the user:
 -How difficult was the app to learn?
 -How much would you pay for this app?
and open-ended questions:
 -What additional functionality would you like to see added to this app?
 -Did anything bother you about this app?

The questions asked of the testers will be submitted anonymously through SurveyMonkey. Those are my suggestions, and now I turn it over to Martin for further comment...


  1. Great to hear you thinking about testing. And a webapp seems like a reasonable choice.

    Personally I would strongly recommend google forms over surveymonkey as being easier, cheaper and more convenient.

    One of the things to keep in mind though is that rather than testing the app, you should test a hypothesis about the app, using two or more alternatives.

    For example, you might say that you believe that one navigation style will be more effective than another, or one data layout might be more appealing than another, and then test that hypothesis with two alternative interfaces.

    With the joke app you could try and do some experimental comparison of how much people laughed at the jokes :-)

  2. I'll have to check out google forms - I haven't ever used them before, but I'm definitely down for cheaper!

    I thought more about testing, and it looks like the display of information is an area where we can use some feedback. More information in this week's post entitled "So Many Choices"