Estibot Wordtracker vs Google Keyword Tool

[nparab]

I bought HdtvReady.info yesterday, but was puzzled by the discrepancy between Estibot Wordtracker count and the Google Keyword Tool search volume figures for the phrase "Hdtv Ready".

Estibot Wordtracker count --> 78/day
If you multiply it by 30, you get ~2400/month

Google Keyword tool search volume figures -->
Approx. search volume, for June: 165,000
Approx. avg. search volume: 135,000/month

There is a huge difference between ~2400/month and 130,000+/month.

What am I missing here? Which figures are more accurate?

Could anyone please explain?

Thank you in advance.

===================================================

I found out why... my mistake, sorry. I had selected "Broad" match-type instead of "Exact" match-type in the Google Keyword Tool.

The figures match when "Exact" match-type is selected.

====================================================

Sorry to restart this, but this time I am having a mismatch with a different domain name. And this time I have checked everything unlike last time.

Search Phrase --> "Vacation Homes"

Wordtracker --> 594/day --> ~18,000/month

Google Keyword Tool:
Avg. search volume for June: 368,000
Approx. avg. search volume: 246,000/month
[Match-type: "Exact"]

Am I missing something again??

 

[conda]

No more accurate, but closely. I love Google Keyword tool

 

[soggyindo]

i hope google's is accurate. i've been a bit suspicious of wordtracker for a while now...

google have the figures (and sell ads based on them), so i'm quietly confident it'll become the most trusted (or should that make us more suspicious?)

 

[-REECE-]

That's interesting...

It looks like the problem is on Google's end. A name getting 100k+ searches per month would definitely have something show up for it's .com counterpart on Alexa and Compete.

I bought HdtvReady.info yesterday, but was puzzled by the discrepancy between Estibot Wordtracker count and the Google Keyword Tool search volume figures for the phrase "Hdtv Ready".

Estibot Wordtracker count --> 78/day
If you multiply it by 30, you get ~2400/month

Google Keyword tool search volume figures -->
Approx. search volume, for June: 165,000
Approx. avg. search volume: 135,000/month

There is a huge difference between ~2400/month and 130,000+/month.

What am I missing here? Which figures are more accurate?

Could anyone please explain?

Thank you in advance.

 

[DomainTalker]

I've noticed this, too...

In researching traffic potential for a given word, there seems to be hugely differing estimates, depending on what you look at...

eg

..Take the word "discounts", for example...


Google's 'Traffic Estimator Sandbox' tool shows 2035 - 2544 est clicks/day (all regions)


Wordtracker shows 390 est searches/day - for the same word.


On a monthly basis, the difference really shows:


Google keyword: 63,085 - 78,864/pm

Wordtracker: 12,090/pm


Be good if anyone could make sense of this.

.

 

[-REECE-]

I just thought of what the reason may be...

Google is probably making it's estimate of how many searches/clicks among both Google Search and the Google Content Network.

I've noticed this, too...

In researching traffic potential for a given word, there seems to be hugely differing estimates, depending on what you look at...

eg

..Take the word "discounts", for example...


Google's 'Traffic Estimator Sandbox' tool shows 2035 - 2544 est clicks/day (all regions)


Wordtracker shows 390 est searches/day - for the same word.


On a monthly basis, the difference really shows:


Google keyword: 63,085 - 78,864/pm

Wordtracker: 12,090/pm


Be good if anyone could make sense of this.

.

 

[dmi]

I trust Google, anyhow.

 

[soggyindo]

I just thought of what the reason may be...

Google is probably making it's estimate of how many searches/clicks among both Google Search and the Google Content Network.

Aha, I had to look up Google Content Network (lol) but that makes sense. If you're doing an ad you want to know how many times you'll get clicked on - not just through google searches, but on sites with advertising too.

Is it absolutely impossible to work out what % of clicks happen on google's search page, vs. through other people's sites, gmail etc?

That way we might have an approximate Google:Wordtracker comparison to work with.

 

[nparab]

I found out why... my stupid mistake, sorry. I had selected the "Broad" match-type instead of "Exact" on Google Keyword Tool.

The "Exact" match-type stats are as follows.

Google Keyword tool search volume figures -->
Approx. search volume, for June: 880
Approx. avg. search volume: 1,300/month

Which kind of matches the Wordtracker figure of 2400/month, because Wordtracker takes data from other search engines too.

 

[shockie]

I had selected the "Broad" match-type instead of "Exact" on Google Keyword Tool.

i was just going to point that out, lol. understanding google's keyword matching options is important - especially if you advertise on adwords so that you don't throw your money away.

also, estibot doesn't own wordtracker.

 

[nparab]

Yes, it was stupid of me :p

Also, I understand Estibot doesn't own Wordtracker, I just wanted to say that I had accessed the WT stats thru Estibot rather than from the WT website. I didn't convey that rightly either... Bad day

 

[soggyindo]

Which kind of matches the Wordtracker figure of 2400/month, because Wordtracker takes data from other search engines too.

What % of searches is Google again?

 

[nparab]

Sorry to restart this, but this time I am having a mismatch with a different domain name. And this time I have checked everything unlike last time.

Search Phrase --> "Vacation Homes"

Wordtracker --> 594/day --> ~18,000/month

Google Keyword Tool:
Avg. search volume for June: 368,000
Approx. avg. search volume: 246,000/month
[Match-type: "Exact"]

Am I missing something again??

 

[Sleepys]

Google's tool shows more results than any of the other tools. I would guess that the other tools may be closer to what we are looking for (not necessarily closer to reality), because often times, Wordtracker, Keyworddiscovery, and Microsoft Adlabs all show similar search estimates. It is Google that often times skews far off the others.

What does this mean for us?

Very little really. It just means that now that Google gives numbers instead of that stupid green bar, we have to device our own system to quantify a name based on their estimate. For example, if you would have normally regged a name that showed 300 monthly searches with keyworddiscovery or wordtracker, maybe you should not register every name that gets 300 searches with Google. Develop your own system for the keyword tools to determine how much weight to lay on each one.

 

[nparab]

Yes, that would be fine if it was consistent... but, in some cases Wordtracker shows significantly higher volumes than Google.

In the case of another (adult) phrase, Google shows ~90K searches (Exact match) and Wordtracker shows ~180K searches

 

[tech4]

Its probably that the data pulled were from searches base off google's popular billion of searches. While that off Estibot are from another search engines which does not use google's data.

 

[Sleepys]

Yes, that would be fine if it was consistent... but, in some cases Wordtracker shows significantly higher volumes than Google.

In the case of another (adult) phrase, Google shows ~90K searches (Exact match) and Wordtracker shows ~180K searches

90k and 180k are both extremely high. The amount can change from month to month, and if it was an exact science, they would all be the same. Results like the one above would give me more confidence then if it showed 80k at one and 10k or less at the other (although those are pretty good too). Keep testing it. Trial and Error. That is why I said develop your own system. I will say that the more consistent the results the better.

 

[james2002]

I found out why... my stupid mistake, sorry. I had selected the "Broad" match-type instead of "Exact" on Google Keyword Tool.

The "Exact" match-type stats are as follows.

Google Keyword tool search volume figures -->
Approx. search volume, for June: 880
Approx. avg. search volume: 1,300/month

Which kind of matches the Wordtracker figure of 2400/month, because Wordtracker takes data from other search engines too.

Exactly. I think you have to trust google.

 

[estibot.com]

estibot doesn't own wordtracker.

Not yet

Anyway, welcome to the perplexing world of search popularity. One thing's certain, and that's uncertainty. Google is probably trustworthy within its own context, but still I have a feeling that it overestimates the search popularity for a domainer's purposes.

You hit a nerve, because I don't think we are ready to answer your question just yet. We are just seeing the data for the first time, too, and we don't yet know exactly what it means.

One problem is, we are used to evaluating domain names by their Overture score. Even though Overture is obsolete as data, it's still used (and mostly useful if you know how to interpret it) and we still tend to use it as a baseline for our evaluation, meaning that if we see 1,000 searches per month, we think it's a decent domain name for a dotcom.

However, Overture 1,000 <> Wordtracker 1,000 <> Google 1,000

They are different for many reasons, one being that none of them try to estimate the internet-wide search popularity, but rather evaluate the popularity within their own network of search engines - as far as I know. Besides, even if they did, they don't really know (well, Google might know, they seem to know everything)

The only way to solve this from a domainer's point of view is pick one system and learn that system's internal logic. Or use them all, but then, learn them all. One has to know what the figures mean. Run a few hundred well-known domain names through e.g. Google Adwords, look at the results, compare the results to the known sales prices of those domains (e.g. from NameBio.com), and compare to the archived Overture data to get a feel of how they were used to be rated, and so on.

Once you learn to interpret the numbers, I'm sure Google's data is good - but it's as yet unproven in the domain name market -> we don't know exactly how it's calculated, and how it relates to the type-in traffic. We already have a good idea of how Overture scores are connected to domain name value, because it's been studied so much, and we know when NOT to trust Overture, too.

My feeling about Wordtracker is that while their data is good, it's far from complete - the database of search terms is small - Overture is better for completeness, and I'm sure Google is the best of the three for that.

I hope Google will release an API access for the Keyword tool, if / when they do, and if I can afford it, I will study it to death and incorporate it into the keyword tools on my site.

-Esa

 

[Sleepys]

That is exactly what I was saying. :tu: