Ian Franklin is a usability testing professional and founder of usability consultancy If Only Consulting.
Web designers and website owners have been practising usability testing for 20 years now, but the majority of us still don’t do it – despite the fact it can deliver incredible benefits and amazing value for money.
Over the years, the methods have got simpler, quicker, and cheaper, to the point where there really is no reason not to do even basic testing.
In this article, Ian addresses the seven most common excuses we all use to avoid doing usability testing.
Myth #1: You need a usability lab and/or expensive equipment
This was true 20 years ago when you needed to hire a lab with two way mirrors and video recording suite that looked like something from Nasa’s Cape Canaveral.
Ten years ago even a ‘lab in a bag’ cost several thousand pounds/dollars and you needed to lug a big camera around.
These days with broadband internet, cheap good quality webcams, direct screen recorders this is no longer the case. Remote testing by a user sitting at home is now possible.
Myth #2: It’s very expensive
In the days of usability labs this was indeed the case; as well as lab hire other costs included accommodating professional staff and users (meals, hotels etc), analysing hours of video tapes, and writing big thick usability reports with many appendices to justify the expense. All done at nice consultancy daily rates.
Nowadays with unmoderated remote tests 10 usability sessions can be collected very cost effectively.
Of course it’s up to you if still want the big thick reports for your CEO to prop their desk up with, but analysis and reporting can be done in a couple of days with annotated video clips supported by bullet pointed recommendations and screen shots in PowerPoint.
Myth #3: You can only test at the end (when the design / development is completed)
The overriding principle of user centred design is that you get feedback from users throughout the whole design process; from concept to post-implementation.
Unmoderated usability testing is very flexible and has a very fast turnaround – so is ideal for Agile development.
You can test navigations and menu labels (card sorts), concept layouts (wireframes), and prototypes (e.g. Axure or Balsamic) as well as fully finished deployed solutions.
You can also test across platforms web, smartphone, tablet, kiosk, console, iDTV etc.
Myth #4: Our website changes so much what is the point?
So this means that either you are very confident about your design guesses about your users or you don’t care if some users occasionally find your designs so unusable and frustrating that they don’t bother to come back ever again.
Remote tests can be turned around in a few days and you don’t need many users to uncover the main issues.
I have seen this on many sites where much of the site was good to OK but the parts which confused users were the ones that stopped conversion.
Myth #5: All users are different so what is the point?
I will give a psychologist’s answer to this and say Yes and No.
Yes in that we are all the product of different circumstances and genetics.
No in that there are certain characteristics where we are remarkably similar: memory capacity, how we read things, how we see pictures, how we notice things, how we group information and follow forms.
All of this similarity comes into play when we look at websites, navigate menus, and look for cues and signposts on what we can do next.
Myth #6: Is the user sample representative?
Statisticians have some nice formula for calculating what a scientifically representative sample is: if your total user group is 100 it is 80 of those people, if it is 1 million it is 384. No one will put 80 people through a usability test, let alone a few hundred. However there is a rule of thumb with usability testing: at least 10 users will bring out the major usability issues and a lot of the subtle ones to. Why – well think of usability testing as more of a case study than a survey; and as the field of medicine shows sometimes a case study of one can tell you an awful lot. Just think what 10 case studies can tell you.
Myth #7: It’s just anecdotal evidence
This one comes from the confusion of the difference between qualitative and quantitative data, and objective and subjective data and that real science involves complicated equipment to get accurate measures to a nano-thingy.
How did Galileo measure the swing of a pendulum when he had no stop watch? He used musicians playing at a set tempo and counted the number of bars. Now you know
this was Galileo… a “real” scientist?
Watching people complete/fail tasks and asking them to rate the usability on a seven point scale gives very powerful data.
And by the way, this is how ergonomists and psychologists do usability studies on nuclear power control rooms, air traffic control systems and airliner cockpits.
If such data is good enough for that it is more than good enough to gauge the usability of an online shopping basket or Smartphone app.