Data Discrepancies – Google’s (Old) Keyword Tool vs. (New) Keyword Planner

Introducing the Keyword Planner

Google AdWords logoBack in May, Google introduced the Keyword Planner: a new version of the Google AdWords Keyword Tool, a handy tool for both PPC and SEO folk for keyword research purposes.

Truthfully, for keyword research projects I’ve carried out recently, I’ve simply continued to use the (old) Keyword Tool. But recently I decided to try out the Keyword Planner, which I’d heard (on the grapevine that is Twitter) wasn’t radically different to what it was replacing, as it was pretty much a merge of the Keyword Tool and the Traffic Estimator and therefore seemingly an aesthetic change that simply combined the two within a new interface.

In order to get properly acquainted with its layout, I decided to run the two tools side-by-side on a quick keyword check job.

A slight discrepancy…

And that’s when I picked up on something alarming…

For effectively the same tool (i.e. one tool replacing the other), the data doesn’t line up.

Looking at “it jobs” keyword suggestions for Computer Recruiter, I noticed discrepancies with both the search volume and average CPC data.

Here’s “it jobs” in the (old) Keyword Tool:

"it jobs" Keyword Tool screenshot(Click to enlarge)

And here’s “it jobs” in the (new) Keyword Planner:

"it jobs" Keyword Planner screenshot(Click to enlarge)

I included all three match types in the former because I wondered if that was the issue (as the KW Planner doesn’t seem to let you choose), but they’re all different anyway. Besides, in the Keyword Planner, if you hover the mouse over the [?] next to “Avg. monthly searches,” you see this:

Keyword Planner search volume info screenshot

From the first paragraph (emphasis added):

The average number of times that people have searched for this exact keyword based on the targeting settings that you’ve selected.

So it should line up with KW Tool’s [exact] match data, but… it doesn’t.

12,100 for one, 14,800 for the other.

Now I’m 99% sure I’ve got all the settings right: same country, same language, etc. Maybe there’s some hidden setting on the Keyword Planner I haven’t found that’s causing the discrepancy. But I’m pretty sure I’ve configured it properly and therefore the data should match.

Lower CPCs – a conspiracy…?

Google money imageIt’s bad enough that the search volume data doesn’t line up. After all, which one is the correct one (if either of them are)?! If I’m conducting SEO keyword research for a client, which data should I use?

But what’s worse is this… For all the keywords I’ve checked, the average CPC is significantly lower in the Keyword Planner. Why is this a big deal? Well, if an advertiser sees a lower CPC for their keyword, they might be more inclined to go for it.

So is this merely a discrepancy between the tools, or a subtle conspiracy by Google to try and encourage a few more people to use AdWords?

Here’s some data for you…

(Note: all the KW Tool CPCs are based on the [exact] match equivalent of each keyword)

“it jobs”

  • KW Tool: £1.28
  • KW Planner: £0.83 (£0.45 less; 35% lower)

“it recruitment”

  • KW Tool: £5.28
  • KW Planner: £3.03 (£2.25 less; 43% lower)

“car insurance”

  • KW Tool: £8.88
  • KW Planner: £5.97 (£2.91 less; 33% lower)

“payday loans”

  • KW Tool: £9.92
  • KW Planner: £6.43 (£3.49 less; 35% lower)

These aren’t just subtle – for the latter few, you’re talking the difference of a few quid. An advertiser may think that £5+ per click is too much to spend, but £3? Ahh heck, why not! They might see different data again when they actually use AdWords, but if it’s all been sorted out with the boss signing it off and a budget assigned to it, the advertiser may continue to give it a go. And Google gets money that they may not necessarily have gotten otherwise…

So what now?

Honestly? I’d love to know what Google have to say about this. If one tool is replacing the other then why does the data differ anyway? Aren’t they fishing from the same pool (so to speak)? Which one is right and which one is wrong? Why do the average CPC amounts vary so much? Hopefully we’ll find out soon enough…

[Google money image credit: Keso S.]

Update: @OldMatt has passed on this support article on AdWords Help where they explain some of the differences.

SEO , , , ,

6 comments


  1. Hey Steve, I missed this from before.

    I think the difference is that the old tool was desktop searches by default, not including mobile, and the new tool is all of the data aggregated together.

    This means higher total exact match searches, and likely lower average estimated CPC’s as mobile CPC’s are often lower.

    Of course what’s interesting is that (and to be fair I haven’t used it to add keywords to new ad groups yet – so this is based on what I’ve read) keywords get added as broad match by default, even though they’re shown in exact match format. So to me it seems that Google are looking to lower expectations whilst still keeping up clicks and competition.

    • Steve

      Cool, thanks for the info, Harvey :-)

      If I’ve understood what you’ve said about broad/exact match correctly then that’s a bit bad… I try to steer people away from using broad, so if Google are trying to ‘trick’ people into using it (so that they can potentially get more clicks, and therefore more money) then that’s not good at all…

      (Sorted out those typos too, as requested…!) ;-)

      • edinufer

        thats great, many people tend to use the broad vs. exact, which anyways wont matter that much know according to this video
        http://goo.gl/z5K873

        also i think it is good they explain more in detail the data values that are coming from the planner tool,

        i wonder what would happen to people who might wanna get the phrase or broad data volume after they completely transitioned to the new system ?

  2. What I’ve noticed is incredibly low traffic numbers that I cannot believe are anywhere near accurate. For example:

    education: 480 searches per month
    management: 260 searches per month

    I hope I’m just doing something wrong in my searches, because those numbers just can’t be right.

  3. The new tool assumes the existence of a web page, while the old tool enabled pages to be created based upon popular search terms (exact match) AND concepts (broad match). It is nothing like the old tool and quite frankly has lost its appeal to me. Now I have to create the page first before establishing the appeal of potential keywords, and can no longer build pages to meet current needs.

    Sure, I can use specialist keyword research software, or search for popular search terms, but Google itself should offer its Adwords clients this type of service. Maybe not free to the general public, but at least to paying advertisers.

    Pete Nisbet

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>