When testing people in front of the computer screen, you need to guide them in some way through the process. The usability goals need to be operationalized in order to check if the product meets them. Operationalization means to take an abstract goal and describe it in terms that can be measured and observed. For example, a site for an airline operator has to be “easy to use”. There is no unit to measure easy to use. You can barely offer somebody the explanation “the site was 3 easytouses”. This has to become something more testable. How about “The site has to be easy for new users to book flights”? Not enough. Having this goal we can break it into little tasks.

How much time does it take for a newcomer to get to the page with the flights? How many clicks does it take for the user to book a flight or find information for a flight? You can invent some more…

But how to ask the right questions when testing if these tasks can be accomplished easily without the user being confused?

The questions must be on point and must be easily related to the tester. What I mean is that the task should vary from user to user. Don’t ask “Try to book a flight from Los Angeles’ LAX to Toronto’s Lester B Pearson on the 23rd of November.”, but “Try to book a flight from your current residence to a holiday location for the weekend”. Tasks must be something a user would do. The service will actually be used by these people, and they certainly won’t try to go someplace and do something pretending they’re somebody else.

Except if you’re targeting a service for stalkers or maniacs

Popularity: 100% [?]