Best,

Breakline

So what is the next best thing? I wish I knew. As I said earlier, there is a real need in the market for this kind of data, and there really isn’t anything else out there — at least, nothing that’s affordable for a small business. You can get statistically sound data from Nielsen//Netratings or perhaps Comscore, but it’s extremely pricey. If I could come up with a easier and more accurate way of getting this information, I’d be doing it.

>>>IMHO, the data being perfect and the data being valuable are two different issues.

You make an excellent point. If you are using the data you get from Compete to identify actionable issues (and better yet, profit from it) then there’s no doubt that it’s valuable for you. But if you’re in a situation where you have to stand behind the numbers you base your decisions on, transparency and accuracy are essential. I’m definitely going to think about this some more. For some of my sites, it may be enough to say, “Yes, the Compete numbers are different, but here’s why — and here are the interesting things we can do with them.”

@catpickett: I think Rae said it best with “… the data being perfect and the data being valuable are two different issues”. You just have to know how to use it and what to look for – hence the review. I tried to focus less on the hard numbers and more on the possibilities and potential.

>>>but they’ve gotta come up with methodology that produces accurate and consistent results

I’m curious as to what you use then? Of course this data isn’t as good as the log files from ebay.com – but I’m certainly not getting those – so what is the next best thing?

For me, in an affordable price range (a few hundred a month) I don’t see much better than Compete available. And the alternative is nothing or free keyword research tools that don’t provide the same depth or ability to search for sites getting traffic based on a keyword etc. I can personally say that in the last two weeks, this data has identified holes we’ve been filling at a profit.

Did you actually try the Pro version? After reading your post, I can’t see how anything other than log files could even come close to what you describe as an unbiased sample set.

IMHO, the data being perfect and the data being valuable are two different issues.

For example, last night, I spent three hours running three reports and came up with 46 new article topics that have search volume according to both compete and the Google keyword tool and a listing of 143 websites targeting our keywords whose backlinks we can now trace. To me? 199 dollars well spent.