The Google Keyword Tool Still Tells Lies

[LiamP]

Here's an example I found last week which serves as a reminder that the Google Keyword Tool can't be trusted 100% and you need to cross-check.

I found dozens and dozens of what looked like great keyword phrases in an established niche. All related to each other, all with about 2000-4000 broad match and about a third of that exact match.

First reason to be suspicious:
Many of their numbers were identical.
i.e. 20 of them had 3900 broad match, 1300 exact match. If the data was accurate, their would be some differentiation

Second reason to be suspicious:
Nobody else is targeting these keywords.
This is one is a bit of a trap as it's easy to use it as a reason to get excited rather than suspicious.

Third reason to be suspicious:
The exact match numbers didn't make 100% sense, as in you couldn't picture that many people typing those phrases in exactly.

So I pasted those keywords into the Google Adwords traffic estimator and it came back with zero traffic for the number one position, even though right alongside it was showing thousands of searches being made .

As a final check I ran a quick two day campaign - zero impressions for 99 keyword phrases which all had thousands of searches a month (apparently).
To check I wasn't making some goof up I did have a related 'control' keyword phrase alongside which Google estimated would get traffic and this did get impressions.

So remember it's a good idea to double check traffic sources before going ahead with work devoted to a new keyword phrase.

[socialbookmark]

You are almost right. Google adwords keywords tool doesn't get right information for many keywords but not for all keywords. I have experienced it for many keywords and the results were true.

[derekjansen]

I always find it handy to cross check those results with a few other keyword tools, as well as check the traffic figures (using SEMrush) of the top 3 position holders of that keyword at the time.

[bfas]

With all due respect, those might be examples, but they aren't "reasons".

I'm not suggesting Google is providing terribly accurate data, but whether you can imagine people typing in a particular phrase, or whether people are targeting them, that doesn't necessarily have anything to do with whether the numbers are accurate or not.

As for the numbers, Google has always rounded the search volume numbers to certain threshold numbers i.e. 1200, 1400, 2400, etc.

bfas

[LiamP]

Originally Posted by bfas

With all due respect, those might be examples, but they aren't "reasons".

I'm not suggesting Google is providing terribly accurate data, but whether you can imagine people typing in a particular phrase, or whether people are targeting them, that doesn't necessarily have anything to do with whether the numbers are accurate or not.

As for the numbers, Google has always rounded the search volume numbers to certain threshold numbers i.e. 1200, 1400, 2400, etc.

bfas

Thanks bfas. I agree that suspicions don't necessarily have anything to do with whether the numbers are accurate or no. They are just that - suspicions, indicators - not reasons or proof. Just an extra hint to dig a little deeper.

As for the numbers, the rounding is not the issue, it's the identical nature of dozens of numbers for different keywords which should be showing different (but rounded) numbers which were suspicious to me.

Fair enough if that's not suspicious to yourself or others.

[LiamP]

Originally Posted by socialbookmark

You are almost right. Google adwords keywords tool doesn't get right information for many keywords but not for all keywords. I have experienced it for many keywords and the results were true.

Thanks socialbookmark, I forgot to put (SOMETIMES) in the title

As you say the results are right most of the time, they just can't be trusted 100% of the time and it is useful to cross check.

[rinor81]

After reading this, what are you thoughts? How to research for keywords besides Google Keyword Tool?

Thanks,

[Osman_M]

@OP: I'm guessing you setup irrelevant ads in your Adwords campaign to test for keyword traffic? What quality score were you receiving for your keywords?

I am having a problem right now. Im using the exact same method to test for keyword traffic but my ad is not showing up for any of the keywords in the list. Adwords keeps telling the "keyword does not trigger any ads."

Do you have any insight?

Any and all help is appreciated.

Regards,
Osman

[LiamP]

Originally Posted by rinor81

After reading this, what are you thoughts? How to research for keywords besides Google Keyword Tool?

Thanks,

Derek Jansen gave an excellent reply earlier in this thread : "I always find it handy to cross check those results with a few other keyword tools, as well as check the traffic figures (using SEMrush) of the top 3 position holders of that keyword at the time."

Also the Google Adwords traffic estimator inside google adwords, its right next to the keyword tool in the menu.

[petercrouch]

Google does not shows exact results and why told accurate? If Google tells you the exact result you will be always on Top and then no need for SEO.

[LiamP]

Originally Posted by Osman_M

@OP: I'm guessing you setup irrelevant ads in your Adwords campaign to test for keyword traffic? What quality score were you receiving for your keywords?

I am having a problem right now. Im using the exact same method to test for keyword traffic but my ad is not showing up for any of the keywords in the list. Adwords keeps telling the "keyword does not trigger any ads."

Do you have any insight?

Any and all help is appreciated.

Regards,
Osman

No it was a relevant ad to a relevant site, the quality scores were good.

Perhaps (I haven't done this before) make a relevant ad to somebody else's relevant site. Set it to $1 a day or something and run it for a day or two. That way you could at least see if there were impressions.
Check Adwords TOS first though, no idea if they allow it and it could get you banned.

[Fernando Veloso]

Sometimes it's more, sometimes it's less then the actual "real" traffic number.

But keep in mind there are keywords with much more traffic than GKT shows you... between a mix of keyword + 2-3 small long tail ones.

Just sayin'...

That's why a small test can and should be done.

For low competition niches, say 2-4K per month exact match, I often do a test with a small site with 10 content pages. I often do this in one day.

IF it does not keep the minimum values I'm looking for I just sell the site later on and get my investment back. Other option is to sell the site to local business owners... these sites can be a pain for us, but a major benefit for them - even with low traffic volume.

Get creative to get your money back...

[sulabh]

I must say looking at your discussion i could understand how much time you give to find out appropriate keyword to help the site, Now coming to post i would google is the only search engine who can provide most accurate keyword then the other keyword finder sites and its up to us what we choose and make that keyword top in in the Search for the Site.