Wordtracker vs. Google- You be the Judge.
July 15, 2008 by Jerry
Filed under Other Stuff, Reviews
I wanted to take this time to clear up some stat issues. Some of the readers of my blog have been questioning my stats, they’re not too sure my stats are accurate. That’s fine, I believe that everyone is entitled to their own opinion.; ) I would also want to make sure that the stats are accurate before I start another niche marketing site. That is why I took the liberty to contact Wordtracker about their stats not matching up with Google’s. This is the email reply I got back from Wordtracker, word for word.
Before I get into the email, If you want to verify that Wordtracker stats are off, do the following;
- Visit the Google Keyword Tool and search “diesel men’s shoes”. You will get an approximate average search of 14,800 for the month of June.
- Got to Wordtracker.com and search “diesel men’s shoes”. You will not find any search results!
Here is the email I got from Wordtracker.com regarding their stats (click to expand).
In this email from Wordtracker states ” We don’t get data directly from Google or any of the engines as they don’t make it available to external tools.” Clearly they have not checked Google’s keyword Tool. lol. So where do they get their data from? Is it accurate? Why am I paying a hefty fee for stats that are NOT generated by the search engines? Then they go on to say “We get our data from metacrawlers, rather than the search engines themselves. After all, the metacrawlers contain the results from the search engines.” HUH? I thought Google uses meta crawlers?
Anyways, to top it off they say “So our data represents around 1% of US search. However, it is worth noting that we go to strenuous lengths to remove all automated or spam traffic. As a result, the search numbers we give you originate from genuine people with genuine credit cards.” Genuine people with genuine credit cards? I don’t get it. This is why there is an issue with the stats. The only question is, who’s stats are you going to trust? Google or Wordtracker?
Now don’t get me wrong, I’m not mad at Wordtracker. I just feel like their stats are irrelevant. Not to mention I have paid for their service only to now find out their stats are off. :\ I’ll just take my chances and follow the big dogs at Google & Yahoo!.
Now, I’m sure you’re happy with Wordtracker’s service and I want you to take this time to share your experiences, both positive and negative.
I look forward to your replies.
If you enjoyed this post, make sure you subscribe to my RSS feed!
It’s a great point you bring up, however, it seems you are not aware that Google only lifted its skirt just last week by providing its search numbers to the public the way it does now. Keyword services like Wordtracker are yet to feel the effects of this.
They are right that this data has been completely unavailable to the public. It is only Google that recently made this move. It remains to be seen if they will continue to allow it too.
I have never been a big fan of Wordtracker. It has only been popularized by the monkey see SEO wannabe community over the years. There have been better data sources that have come along over the years. Wordze has really impressed me recently.
I think the whole keyword data industry will start to see the effects of this move by Google in the coming months. Assuming once again that Google keeps it up. They might be doing this just as yet another experiment.
It is not clear also that the data provided by Google does or does not include scrapper and spammer queries. It would not surprise me if they simply left it in there to inflate the numbers. You would be shocked how many search queries are done by scrappers and auto blogs.
All in all… the only keyword data you can trust, is the data you collect on your own from your own sites. The rest is all tainted, so don’t let anybody make you think that one source is better than any other. They are all still ‘best guesses’.
You have a fine blog here btw… congrats. :)
[Reply to this comment]
It’s great that you’ve followed this up Jerry. I commend you for it!
Great blog.
Going by the posts so far.. It seems you like your sport ;)
[Reply to this comment]
This is so interesting, because tonight I was sitting and doing some research and I was fiddling with the new feature on Google Adwords KW tool that shows the search volume numbers and I decided to compare those results with the free Wordtracker and they were not even close!!!
I completely agree with you, and I had done the comparison’s before and they seldom match or even come close, now this also applies to some of the free KW tools as well, including SEOBook’s free tool, all though that one seems to be more close to Google results.
That email is very lame sounding, and considering the money they charge for their service, it’s kind of appalling to find such discrepancies.
I am a KW Discovery user (paid service) myself and I think I am going to do some comparisons to make sure that I am getting what I pay for.
Thanks for the interesting and honest post!
JR
[Reply to this comment]
JR,
Thanks for the comment. Hey, let me know how those search numbers go for KW Discovery. I’m curious.
Jerry
[Reply to this comment]
Well, the “truth hurts” saying has never been more true than today. On the bright side I will be saving some bucks as soon as I cancel Keyword Discovery.
Keyword Discovery results are completely off from what shows in the new monthly search numbers feature in the Google Adwords Tool.
I searched about 20 different keywords and not one was even close. What is more concerning is that KW Discovery allows you to pick from several databases, one of which is Google and even THAT one was completely off.
I started to think that maybe Google was off, in hopes of saving at least some of the regret I was feeling in spending the money these last couple of months, so I picked some high-end keywords.
For example, “laptop”, this should have a very high search volume, well in Google it did, in the millions per month, but KW Discovery showed up in the thousands and that cannot be accurate.
I also compared several of the KW with SEOBook Free Tool and actually they were the closest to Google, all though not as close as it should be, several of the counts came in way under.
So, there you have it!
Now I tend to agree with Jonathan that the Google count may be inflated, with spammers and auto search included, however, I think paying such a high price for some of these keyword tools is still concerning, plus the fact that the Wordtracker email you got Jerry does not clearly define how it is that they are getting their results.
I think this is going to stir up the Keyword Tool Biz in the weeks to come, this is the first blog post I have seen about it, but I doubt it will be the last. I may just do a post myself with a link here, of course I am an affiliate of a couple of these keyword tools and since I never recommend stuff I don’t use and don’t find effective I will have to find new stuff to advertise :-(.
Thanks,
JR
[Reply to this comment]
Great post. I became skeptical, and stopped using, paid KW reasearch tools long ago. I began comparing my results, based on log entries, to GTrends and other tools. GTrends always won by a long shot. Now the new numbers available from Google put a spotlight on just how off these tools are.
I think the release of these numbers will have a powerful negative effect on the paid KW research industry.
So what are your thoughts on Google Suggest? Do you think it will compress the long tail?
Thanks for a great blog.
[Reply to this comment]
I’m really concerned with some of these comments. You guys are going to shoot yourself in the foot.
I’m really afraid you guys are missing the point I was trying to make about trashing keyword tools in favour of using googles… Others who are more experienced in the industry see this, but people new to it don’t.
Here is someone else who has had real life experience in comparing the Google Adwords tools to other services when they used it for SEO… its very in depth and will help you understand why this would be a very bad move:
It was posted at Smackdown!
Hope this helps
[Reply to this comment]
Hi Jonathan,
I understand what you are saying, but what I am saying is that there is no way that laptops could only have 3,000 avg searches a month, NO WAY and that is what Keyword Discovery showed, AND it was their “global database”. And, I never said that google’s numbers were precise, the point was that to pay a hefty monthly fee to some of these keyword tools that are showing really off numbers is concerning.
I also said that SEOBook Free Tool seemed to have the most viable numbers of all the non-google tools and it is free.
And, I agree with you that these tools are only suggestive and you will not really know till you test it, this applies to all tools, and I use a combination of research techniques, not just keyword tools to search for niche/product popularity, but the whole point is if I am paying for a keyword service, then I want more accuracy then what is evidently being generated.
And, what happened to the guy who wrote the article on Smackdown, also happened to a webmaster I know, who used a combination of Wordze and Keyword Discovery, where he found a low competition keyword that showed over 5,000 searches a month, he built the site, sits #1 on Page 1 of Google for two months and has had a total of 10 clicks, 5 PER month!!
So, if this were to happen to me, I certainly would feel better that this happened with a free keyword tool than with one I pay good money for every single month.
That’s just my 2 cents.
[Reply to this comment]
A couple of things from the post to keep in mind is that Google is NOT a meta crawler and I may be wrong but I don’t believe they ever have been. I believe the actual definition of a meta crawler is a search engine that queries the databases of other search engines to produce their results, they don’t have their own database.
Additionally, simply because Google now shows search volume in their Free Keyword tool doesn’t necessarily mean they’re allowing access to external tools - like WordTracker - to parse and analyze the data.
In answer to : “who’s stats are you going to trust? Google or Wordtracker?” For myself, I’m going to continue using a cross section of all the data available as Google doesn’t represent the searches made at the meta crawlers or MSN or Yahoo and WordTracker doesn’t have Googles or MSN’s or Yahoo’s search results.
With all of that said, no one source is going to have the exact same data as another source.
I think maybe the biggest question is what types of filters has Google applied to the search volume numbers they publish.
At least, that’s the way I see it :)
BTW Great Site!!
[Reply to this comment]
One thing to remember when doing keyword research is that not any one single tool will do the job as their numbers come from different places.
For instance, Google’s numbers come directly from their database. They aren’t a metacrawler search engine.
Metacrawlers are search engines that don’t have their own database and cull search results from other search engines’ databases.
So when comparing Wordtracker results to Google results, of course, there will be different numbers as they get their information from completely different sources. Google from their own database, WordTracker from metacrawler search engine databases.
[Reply to this comment]
I just had to do a search for ‘wordtracker accuracy’ after comparing googles numbers to the paid for service of WordtrackersI who I emailed and had pretty much the same puzzling response.
I wonder what I have been paying for with WT when ‘Osmo’ with 9,900 searches on Google, which sounds realistic gets just 77 with Wordtracker. I would rather use the results of the majority of searches, not the minority.
[Reply to this comment]
Be careful. NOBODY is displaying accurate numbers. You have to test your results before putting serious time into them. Google’s keyword tool is ok. Wordtracker has the advantage of offering more keywords then google per search, but google does offer values on adwords which gives some idea of the ability to monitize on those keywords. I’ve tested from all the major keyword packages and none stand out, not even google. You always have to test. It’s a simple matter of creating an adwords campaign and running a simple ad and only allow listings with the google search engine and NOT the content network. This will offer you be best idea of results.
Frank.
[Reply to this comment]