It's true that mere searches for Google instant might not directly be a measure of it's success. But most certainly, it is a measure of the level of interest sustained.

a)Even if the phrase "Google instant" were to be done using Google Instant, wouldn't Google have done something to incorporate it in the trends? (User clicks on links, right?)

b)Google Instant was not released to the general public all at one go.It was done in phases. So if it really did have a viral effect, and sustained interest, the graph would not be a spike and a fall-off.

Take the case of Google wave- Here 's the Google trend of October 2009, (It was released on Sep 30)

http://www.google.com/trends?q=google+wave&ctab=0&ge...

Compare the news reference volume of this and Google instant. Google instant seems to have a sharper fall in news reference volume than even Google wave! This has to be linked in some way to the number of people using/ enjoying Google instant! My hunch is that, there haven't been as many takers for Instant, as Google might have wanted-- at least,not yet.