Nothing is going to really give you any thing that is all that close to dead accurate.

If you are trying to really hammer down local search volumes within Googles Index, I would highly suggest you take a once over with Google Insights.

Searches or Impression really do not matter all that much overall, the fact is people search for damn near everything today, as long as you are the most interesting and relevant result that speaks to the end user, you will earn the Click Throughs which at the end of the day is typically all that matters.

I'm not asking for perfect accuracy, I'm just trying to find out why there's such a big difference between spyfu (12k) and GAKT or MNF (2.4k). It isn't a ~10-20% gap, but more like 500%, which is really significant. I'm more of a "number" person, and what you're saying doesn't make sense. A keyword that gets 1 search daily will have no traffic whatsoever even if your site is GODLIKE interesting. More searches, more traffic, more clicks, more money; and that's where keyword research takes place.

Please check with sktool (Google). This will give you more Insight. thanks

If you really are a number person than do what all of us do and Run targeted PPC ads for the terms in whatever area you are targeting for 5 days and split test everything.

Thats how you can determine actual costs.

No keyword tool actually is trying to be all that accurate, its not really what they are all about from a programmatic point of view...

I can only assume you are fairly new to all this so take it slow and learn as much as you can about the real world difference by trying and testing.

Most of us have been doing this for years and we still learn new things everyday.

Just use those numbers to validate the results of your research and assume the smaller is the right one (in your case 2400 searches is not bad).