Pattern 8: Get-it? **
AKA:

You are testing you web site for usability and have completed AUTOMATED TESTING (6) and may having begun USABILITY TESTING (7). However...

Full usability testing using with large numbers of users or in laboratory conditions can be very expensive and time consuming: thus affecting timely update to the site. However, we must only deliver tested, usable sites.

Therefore

Use anyone available as testers if real users are not available. Explain nothing in advance. Ask them: ‘Do you get it?’ Ask other suitable open questions, perhaps based on the list above. Allow testers to comment freely. Use get-it? throughout the life-cycle from planning to the first released version. If testing is good, do it all the time. If possible, use some people who have never used the web before. Let then access the site from home as well as your offices. Offer them a reasonable payment for their time. Keep the test sessions down to less than an hour.

This pattern is terminal within this language. However, make sure you also do USABILITY TESTING (7) in parallel with this pattern.

Sources
Krug (2000), Tognazzini (1992).

This is a very simple pattern is taken from Krug (2000). He suggests taking almost anyone that happens to be around, preferably the same kinds of people that might be users, and asking them to sit down and use the new, partly tested site. Then just ask them if they get it.

Remember that the best testers are not always the most co-operative personalities. The best ones are the cussed, impatient ones who are just too busy to play. Remember too that developers always want to play!
The sort of responses to be interested in include inter alia statements about:

• the value proposition;
• what the organization does;
• what makes them different from other organizations in the same or similar businesses;
• what is the purpose of the site;
• what it contains;
• how it is organized
• what could someone gain from a visit;
• was the experience a pleasant one;
• how this site compares with other sites;
• the look and feel.

Krug distinguishes get-it? tests from key-task tests, which are based on the use cases that we know about during USABILITY TESTING (7). We should also distinguish them from the regression tests carried out when we AUTOMATE TESTING (6) when content is updated and tests across different browsers including TWO-YEAR OLD BROWSERs (10).
The cost of this kind of test is not prohibitive (cf. Tognazzini, 1992). Therefore you can afford to do it frequently. The payback is usually well worth it.

As with usability tests do this early in the project and continuously throughout it. If tests are performed at the end then it will be too late to fix problems before the scheduled release date.

Browse the language	What is Wu?	Look at an example pattern sequence	Structure of the patterns
Comment on Wu	Contributors	Return to TriReme home page	Links to related sites