Everyone is talking about the new Google Trends for websites – Graywolf has a piece on it, as does Aaron (where this post started as a comment. Hi Aaron! Look! Blogging!). But I think its important to note that, while people are complaining that this is giving away the farm, its not as bad as it looks at a glance.

First, have a look at some results:
http://trends.google.com/websites?q=yahoo.com&geo=all&date=all&sort=0
Done? Good.

One of the things people are worrying about is that their keywords are being given away.
The column headed “Also searched for” would appear to be *other things the user searched for*, NOT *the keywords used to reach this site*
If I am interpreting the label correctly, it is a completely different critter than most people are worrying about.

Secondly, we’ve always been dubious about putting GA on money sites haven’t we? And here we seem justified – except… it says “aggregated opt-in anonymous Google Analytics data”. Note the “opt-in”. My data’s not up there, because I didn’t opt in when offered.

I’m not really sure how valuable this info is at all, apart from a general ooh-ahh :)
I looked at the report for one of my sites and saw one of my major competitors there as another site my visitors visit. Yes! I thought, raw data on the competition! But… the mathematics doesn’t agree that its useful. An unscaled graph comparing the two doesn’t really tell you anything.
Even if I compare my traffic to theirs using a base line of a 3rd site which visitors to both our sites visit I still don’t have a valid baseline, because I can’t assume our two (overlapping) pools of visitors both visit site 3 at the same frequency.
So….
Nice try on Google’s part, but its really only for fun. :(