If you haven't seen it, Google Trends allows you to see how often people are searching for a term and where they are located. Another useful application of it is the ability compare different search terms. In doing this you can compare the usage of different terms to describe the same thing. Why is this useful?
Often we sit around trying to think labels for navigational elements or sections of a website. Sometimes we argue about how the users relate to language that is familiar to us but less so for them. Being able to see what people are using to search for our site or how they are combining terms and where they are searching from gives us an insight into how they are using our website.
Of course, this should be done with the logs from the search engine on your own site as a matter of course. Not only in terms of language but also to guage the usefulness of content, getting an idea about when to push some content and when to pull back on other stuff. (Also, to make sure your search engine works). The academic year is a pretty good indicator of what is going to be required when on a university website but the analysis of search logs can help refine content reuqirements.
Using Google Trends allows you to broaden the user base and see what people are using before they get to your site, not just after they have found you.
A very basic example:
The University of Sydney is called several different names, according to who is speaking. Its full and correct title is "The University of Sydney". Often is called "Sydney University" or "Sydney Uni" or even just "Sydney" if the context is clear. Common sense would say that "Sydney University" would be most used within the general public, simply because it's easier to say. Google Trends bears this out. In general search it is far more favoured than the correct title. Surprisingly, "Sydney Uni" was used much less. The city in which it is most popular is Sydney itself, which points to 'local knowledge' playing a part here.
In marketing and communication terms, you can see how different geographical areas use different names for the insitution and are possibly more likely to identify with that which is familiar to them. A case could be made for shortening the name to "Sydney Uni" for some campaigns when the context favoured it.
You can, of course, also check out the opposition...whether it be by the more formal titles or the shortened versions of names.
(To my friends at UTS and UNSW: this was purely an exercise in illustration!)
Admittedly, Google Trends isn't going to give you a spot-on analysis of how people are thinking, but it is another tool to add to the collection.