Interaction Design Day: Usability Bloopers

Interaction Design Day, Copenhagen
I’m here live blogging at the Danish Interaction Design Day 2007 in Copenhagen (link in Danish). This post is probably the only one I’ll live blog today, as I only attend part of the conference.

The mini conference is a fine mix between trade show and presentations. Best thing is to meet many of my colleages in companies I have worked with. Thomas and his colleagues from, Rolf Molich from DialogDesign, Nanna, Christina and Jacob Hage from Creuna, Anette, Hannah, Linda from KMD, and many more, both project participants and clients.

Ben Schneiderman, Amsterdam, 2000

Unfortunately, I’ll miss Ben Schneiderman‘s keynote. Actually I saw him 7 years ago (when starting my usability career) at a usability seminar in Amsterdam. He was a very inspiring speaker. I remember he spoke a lot about visualizing large amounts of data, and gave some examples which I have found useful at lots occations.

One of the examples was about using SpotFire (where he is on the board), visualizing all homes being sold in the Washington DC area. Using sliders to filter the result set, it was able to visualize meaningful connections in the data. Such as where are the finest or worst areas of the city, visualize where it would be optimal to live if you want to live less than 2 miles from a school and 4 miles from work. Those were eye-opening examples (and I actually saw recently that offers something very similar to what Ben originally demoed in Amsterdam. If you are able to, then don’t cheat yourself for hearing him speak. (and please give me an update afterwards or in the comments :).

Rolf Molich: Usability Bloopers

Resume of Rolf Molichs talk on Usability bloopers. Which I live blogged. See my notes here.

Usability bloopers is not only about how we bother peaceful peoples websites. But it is more an angle at how we as usability experts fall in the various usability traps.

How do you become a usability expert? Back to that after examples.


Example 1: questionnarie

Example 2: “How likely are you t obtain a mini-statement when yu visit the ATM or SST?”
Problems: WHat is a mini-statement? What is an SST? There is no option to answer “I don’t understand the question”

So, we need to take our own medicine. These questionnaires would never slip through a user test.

Test tasks

Example: “Go to and find the phone with FM radio that suits you best.” (in the Danish version there is a spelling error which is a minor flaw).
Problem. The task is closed: You make them pretend that they are interested in a FM radio.

Resolution: Make the task more open and motivative: “Here are 250 USD. Go out and find the mobile phone that is best for you”.

Example 2: Test of “Find out what the weather will be in Fyn, Denmark tomorrow”, “Find what is on TV3 channel thursday afternoon at 5PM”, “Find out who won 2nd place in Post Danmark bicycle race”, “Who is the economical responsible for DR (Danish Broadcasting Corp”.

Problems: Amazingly few tasks. And less relevant for ordinary Danes.
Resolution: More relevant task for ordinary Danes. Tasks like “What is on the largest radio channel this afternoon”, “What will the weather be in the area that is relevant to you”. Also, make more tasks.

Again, open up for the user and what motivates the user. Find out what

Usatility test: Reporting

“I thought it should be obvious that reports should be usable”

The example text is from a usability company’s website. The marketing text lacks a link to an example usabiilty report.

Resolution: We preach to give examples, but there is missing an example on this report.

Example from a test report from IKEA. A group consisting of 11 women and only one male (which is nurse — traditional womens job, by the way).
One of the participants has family relatins “Married, x children”, which looks like sloppyness.
Lacks information about to primary pieces of information:
* How often are you at an IKEA store
* How often do you use the IKEA website

One is “student”, but in what area. Another one is “Project manager”. But in what area.

Expert reviews

Expert reviews should be created by an expert. I have tried to develop an expertometer to clarify if people are experts.

Example Danish Red Cross homepage. (where looking at the top right zoomed in quadrant)
Problems with newsletters. The expert must know important issues on newsletter subscriptions.

Conclusions: Rules for quality assurance. My studies af professional usability people “experts”, show that high quality unfortunately are with fatal errors. Not only occational but more often, most of the results, Rolf sees almost always fatal flaws in groups that are asked to make comparable website evaluations.

Improvement suggestions are often wrong. Existing certification by a firm in England does not really live up to the quality that should be expected from a usability expert (to put it mild and diplomatic). There is lot of negative talk about this in American user forums, and also examples of highly skilled usability professionals that have almost flunked the test.

Technorati Tags: , , , , , , , , ,

One Response to “Interaction Design Day: Usability Bloopers”

  1. Thomas Visby Snitker Says:

    What surprised me the most in Rolf’s presentation was the fact that he actually got a hold of these bloopers. Probably from the same people who made the research them selves.
    This is amazing and very promising for the future; could we share more of our work please? And learn and improve the field of usability testing?
    Then bloopers become valuable assests. (What’s the saying? “you need to fail in order to improve”)