Question for the group:
Learned opinion is that WordTracker is far more accurate and precise than Overture. There are clearly times when OVT shows figures that are just nuts, so I can buy that OVT is often inaccurate.
I spent a good amount of time playing around with Word Tracker today and I am finding that there is just as much oddness in their figures, most of it undercounting:
Example 1:
When you look for variants of the word "travel", the #4 search term is "travel manila". Um. I doubt that.
Example 2:
I have an operating business that I advertise heavily using google adwords on a generic consumer search term. (I would prefer not to reveal the term).
When I check this term on Wordtracker, it gives me a predict count of 380 worldwide searches per day.
My actual search impressions, just from google, on this term are 2,300 (search only, not content partners). Note: this is not predicted data like OVT, this is the actual google search data.
In fact, my click-throughs on this term are 70+ per day. While I am flattered to think that my ad clickthroughs account for 20% of the worldwide search volume, I doubt it.
Example 3:
Finally, a silly, personal anecdote. I live in NY. I order food a lot. I personally probably google the word "food delivery" once a week. Word Tracker predicts 167 searches, daily, globally for this term. Again, I doubt it.
My guess is that midtown manhattan between 6 and 9pm tops that on a daily basis. Granted, this one is not a fact-based assessment.
Thoughts and Hypotheses
I would welcome experience and thoughts from others.
My hypothesis the type of person that uses a metacrawler to search (which is how they get their data) is substantially skewed from the type of user that uses Yahoo, Google, etc and is likely to be male, young and geeky.
That is why certain terms (e.g sex-oriented terms) score very highly on WordTracker but a lot of consumer product terms seem to be lower.
Learned opinion is that WordTracker is far more accurate and precise than Overture. There are clearly times when OVT shows figures that are just nuts, so I can buy that OVT is often inaccurate.
I spent a good amount of time playing around with Word Tracker today and I am finding that there is just as much oddness in their figures, most of it undercounting:
Example 1:
When you look for variants of the word "travel", the #4 search term is "travel manila". Um. I doubt that.
Example 2:
I have an operating business that I advertise heavily using google adwords on a generic consumer search term. (I would prefer not to reveal the term).
When I check this term on Wordtracker, it gives me a predict count of 380 worldwide searches per day.
My actual search impressions, just from google, on this term are 2,300 (search only, not content partners). Note: this is not predicted data like OVT, this is the actual google search data.
In fact, my click-throughs on this term are 70+ per day. While I am flattered to think that my ad clickthroughs account for 20% of the worldwide search volume, I doubt it.
Example 3:
Finally, a silly, personal anecdote. I live in NY. I order food a lot. I personally probably google the word "food delivery" once a week. Word Tracker predicts 167 searches, daily, globally for this term. Again, I doubt it.
My guess is that midtown manhattan between 6 and 9pm tops that on a daily basis. Granted, this one is not a fact-based assessment.
Thoughts and Hypotheses
I would welcome experience and thoughts from others.
My hypothesis the type of person that uses a metacrawler to search (which is how they get their data) is substantially skewed from the type of user that uses Yahoo, Google, etc and is likely to be male, young and geeky.
That is why certain terms (e.g sex-oriented terms) score very highly on WordTracker but a lot of consumer product terms seem to be lower.
