In my view nearly all user feedback can be useful in one way or another. User feedback of the kind you’re describing is certainly subjective on a person-by-person basis, but that’s ok. As the number of feedback contributors increases it becomes possible to spot trends and similarities. That eliminates much of the subjectivity and transforms a random group of opinions into a usable set of data. In other words, if many people tell you that (for example) they didn’t notice the help link and would like it to be larger, it doesn’t really matter that each person’s opinion is subjective. In the bigger picture you’ve spotted a useful trend and can act on it.

One way to give structure to user feedback is to enable people to provide general, “open ended” comments as well as more constrained feedback via questionnaires that address the points you feel are most important. That way even before you start to spot trends in the general comments you can see statistically how users feel about the site’s effectiveness in certain categories.

One other thought before this becomes an article all by itself

I usually advocate enabling users to submit feedback anonymously, and to create very few required fields on a feedback form. That makes it quicker for someone to share their opinion (which means they’ll be more inclined to do so). It also reduces the chances they’ll feel awkward about sharing a critique or concern.

We’ve been through several iterations of our dealer locator, yet usability issues persist. We do not have the resources for extensive user testing. Instead we’ve learned a great deal by allowing users to comment via a link on the locator page. The insights have been eye-opening, and will continue to shape future refinements to the locator.

I would be interested in the author’s and others’ thoughts on using feedback links to collect usability information. Is this information useful or too subjective?