How do you compare year-over-year keyword analysis without (Not Provided) screwing everything up?

[tatermarketing]

We have a seasonal business.  It is of no use to compare this month to last month or even this week to last week.

The only real benchmark is to compare the same time period a year ago.

I'd like to know how well I'm doing for a certain keyword.  If I compare this month's performance to the same month's performance a year ago, I get corrupt results as much is now hidden in (Not Provided).

Any solutions?

[tatermarketing]

Thank you for verifying ... I didn't think there was a great answer.

Google should provide a "(Not Provided)" eg. "logged into Google" segment.  This way, at least I could compare apples to apples from a year ago.

Even then though, the number of users logged into Google this year must be higher than last year so it is still very dirty data ... just less dirty.

[iPullRank]

Hey Ryan,

I would suggest looking at search as a holistic channel rather than reporting YoY per keyword. If that's not sufficient I would suggest you check out Ian Laurie's method for getting your keywords back.

-Mike