A report on the Webcredible site should give us food for thought, if we hadn’t thought about it already. Even more so, it possibly explains a lot of user behaviour across different channels, so there is no need to employ it in the single context of the web.
Some of the key user experiences described in the report – word-blindness, tunnel-vision, impatience and habitual behaviour, all lead to unexpected user performance when using web sites. So what do you do? Obviously there is a need to get it tested by real people and allow real people to feedback any concerns.
Where does this apply to other channels? Get any forms proof read by potential users and employ potential users to test telephone messages or processes. A recent rail journey with a blind consultant confirmed this in other ways, Graham preferred to use the term “usability” testing rather than “accessibility”, since he felt that if a blind person like himself could find their way around a service, most should be able to. So lets have more usability testing on services and processes.