Wordtracker vs. Overture - how do you explain this?

Weird discrepancies ...

[limitup]

Overture data is based on significantly more data than Wordtracker, which explains why Overture results are usually higher than Wordtracker for the same term.

However, quite often it's the other way around - which doesn't make much sense at all.

For example right now I'm looking at one term that has Overture of about 80,000/mo. and Wordtracker is almost 250,000/mo. I come across terms like this every day.

Does anyone have any theories to explain how this is even possible? Or is it a glitch with one or the other? If so, any idea how to tell which is the more accurate number? When in doubt do you just rely on Overture?

[annej]

I've wondered the same. It's unreal how different they can be. In the end I just look at both and use my best judgement about what keywords I want to use.

The highest visited keywords aren't always the best anyway unless they are mixed with other more specific key words.

But the key word tools do help along with a lot of other considerations.

[Mistra]

Of course the data of Wordtracker differs from Overture. Wordtracker claims to collect their data from metasearch engines.

[mona]

Overture's results also include bot searches, not just human searches. Every part of keyword research is crucial. The more data you have, the better your results will be. Here are some more ways to go about it.

1. Use WT, Overture, and Google suggestion tools. Look at all three and try to find a middle ground.

2. Use AdWords campaign tools or set up an Adwords campaign. If you want to go the cheap route, you can set up a campaign and Google will show you the estimated click-throughs for your keyword phrases. Look at the data, but don't set up the campaign. If you're willing to spend a few bucks, do the same but set the campaign up. You'll get much more data this way. Most importantly impressions - the number of times a keyword phrase was searched. The longer you run the campaign and/or the bigger the budget the more accurate your data will be.

3. Web stats. Check your logs and see which keyword phrases are actually being used to find your site.

4. Study the competition. What KW phrases are the leaders within your industry/topics targeting?

If you use all of four of these, you'll have a much clearer idea of which keyword phrases to go after.

[martinibuster]

Overture results also include the aggregates of singular, plural, and misspellings. The only time I see those broken out is when there are significant amounts.

Then you have to guesstimate the amount of queries that are performed by webmasters checking their rankings.

Wordtracker, as had been mentioned before, gets it's information from meta search, but then (last time I looked) it extrapolates and estimates the amount of searches at the various search engines.

Another way to get an estimate of searches is to run an AdWords campaign set to search only, without the partners, and set it full throttle for a day or two- or a week if you can afford it. You can break out your searches according singular, plural, exact and phrase match. Then use the impression numbers for decision making. ;)