Google’s “search volume” estimates are not what you think

Google's Search Volume Estimates can be deceiving for SEOGoogle recently clarified its estimates for organic search traffic, and in doing so, it may just have pulled the rug from under many search-engine-optimization (SEO) programs. Apparently, the Search Volumes presented by Google’s Adwords Keyword Tool, used by many SEO “specialists” to estimate the size of the organic search market and justify their SEO programs, can overestimate the actual organic search market by a factor of 100!

For example, if you thought that the monthly search volume for a specific keyword phrase was 28,000 searches a month based on this Adwords keyword tool, you might just find the actual search volume is 280: a huge difference in market size!

The implication for marketing managers everywhere is that their return-on-marketing-investment (ROMI) calculations used to justify SEO programs could be deeply flawed.

I’ll get to the details in a minute, but first some background for those marketers not steeped in the world of SEO. (You might also want to check out our Glossary of SEO and SEM terms)

The importance of Search Volume and Search Volume Tools

Most Inbound Marketers agree that SEO is the most cost-effective long term investment to drive quality, relevant traffic to your website. More so than the other primary drivers of inbound traffic, namely Pay-per-Click advertising (PPC), and Social-Media-Marketing (SMM).  To justify their investment in SEO programs, SEO specialists around the world try to quantify the number of prospects a well-optimized website can attract.

To arrive at that number, SEO pros use a variety of tools to estimate search volume. One of the tools frequently used by SEO specialists is the Google Adwords Keyword Tool (GAKT).  As an example, see the following report for the search phrase “environmental equipment”:

Google's Adwords Keyword Tool and Search Volume Estimates

Result for the keyword phrase "environmental equipment" using Google's Adwords Keyword Tool

Note that the estimated “Local Search Volume” for the term “environmental equipment”, in April 2010, across the United States was 12,100 searches.

Now SEO professionals have know for some time, based on anecdotal evidence, that the search volume estimates produced by this Google Adwords Keyword Tool seemed a little high. But in the absence of better tools (and remember that Google represents 70%+ of the global search industry), these numbers are nonetheless used regularly to flaunt the size of the search market and justify investment in search engine optimization programs.

Google’s new insight on Search

Just over a year ago, Google introduced a beta of the new Search-based Keyword Tool (SKTool) with relatively little fanfare.  Initially targeted at some special applications, the SKTool did not replace the traditional Google Adwords Keyword Tool as a primary tool for determining search volumes. However, it was noted by SEO pros everywhere that the search volume estimates by SKTool were often drastically lower than the traditional Adwords Keyword Tool.

For example, using the same search phrase of “environmental equipment” above, the SKTool returns much, much smaller search volume estimates:

Google's Search-based Keyword Tool and Search Volume estimates for SEO

Results for keyword phrase "environmental equipment" using Google's Search-based Keyword Tool

That’s a drop from 12,100 searches a month to 155 searches a month!

Initially, Google’s explanation of the difference in search volume estimates between GAKT and SKTool was quite cryptic:

How does the Search-based Keyword Tool differ from the Keyword Tool in AdWords?

Answer: “The main difference between the Search-based Keyword Tool and the Keyword Tool currently in AdWords is that the former generates keyword ideas based on your website, and identifies those currently not being used in your AdWords account. Additionally, the Search-based Keyword Tool provides more detailed data for each keyword, such as category information, suggested bid that may place the ad in the top three spots of a search results page, and ad/search share. Both tools, however, offer the option of browsing all keywords across all categories”

You may also notice that some of the data (like such as the monthly search volume) may vary slightly between the two tools, which is due to different methods of calculation at this time (emphasis mine).

And the situation remained cryptic until Google very recently added this explanation to clarify the situation:

Why do search volume statistics vary between keyword tools?

Answer: “The updated Keyword Tool combines search data across Google.com and all affiliated search properties. Data from the Search-based Keyword Tool is based solely on searches on Google.com (emphasis mine). Numeric data between the tools may also differ due to rounding.”

HOLD THE PRESS! The full implications of that last sentence are that the search volume estimates that many have used to justify SEO programs are not based on Search Engine queries at all, but rather primarily on click-through rates on PPC ads on Google’s advertising network! There may be 11,945 (12,100 minus 155) click-throughs on various ads associated (loosely) with the keyword “environmental equipment”, but only 155 actual search engine queries. Thus the entire SEO potential for “environmental equipment” is 155 searches a month (US), not 12,100.

More practical examples

In case you think the above example using “environmental equipment” is a statistical outlier, here are a couple of other B2B search phrase examples:

Search Phrase

Monthly Search Volume (US – April 2010) based on Google’s Adwords Keyword Tool

Monthly Search Volume (US- April 2010) based on Google’s Search-based Keyword Tool

injection molding

201,000

3,500

marketing automation

74,000

660

fleet management software

74,000

820

Is SEO dead?

What does this mean to the $10+ billion global SEO industry?

In my view, this is not the death knell for the SEO business.  Achieving the #1 Search Engine Results Page (SERP) position in any of the above keywords will still generate a lot of relevant, valuable traffic to your website.

But it is a big reality check for the world of SEO, and will change every ROMI calculation used to justify an SEO campaign. And with Social Media (aka the “new search”) coming on strong, perhaps B2B companies need to take a closer look at Social Media Marketing as a driver of highly relevant traffic, in addition to SEO.

A reminder: A tool is only as good as the person using it

This post also serves as an important reminder of an issue we continue to emphasize at Gossamar.  Tools, no matter how cool, are useless in the wrong hands. Whether it is SEO traffic estimation tool, or marketing automation tools, a tool without expertise and a sound process behind it will deliver flawed results.

Unless you think the difference between 12,100 and 155 is minor.

Until next time – Axel

  • The Gossamar ROMI Calculator
  • Tags: , , ,

    June 4th, 10
    Axel
    Filed Under Blog Posts

    21 Responses to “Google’s “search volume” estimates are not what you think”

    1. Axel;
      What a brave and honest post! As SEO practioners ourselves, our clients deserve to know accurately what the potential upside for engaging us is. And this will help set the record straight. Well done!

    2. [...] This post was mentioned on Twitter by Jay Goldman, Gossamar. Gossamar said: SEO Industry wake-up call. The numbers are not what you think. Read Gossamar's latest blog for the truth: http://bit.ly/8ZFGBp [...]

    3. Thanks for the post Axel it is definitely food for thought.

      I have for some time been dubious of the statistics provided by Google, for example comparing Site Links from Google compared to Yahoo’s Link Domain delivers substantial differences in results.

      Personally, I would not base any site traffic estimates on monthly search volumes. I use them only for comparison between keywords to assist with keyword selection. If you want to take this a step further along your path when you compare site traffic for sites obtaining visibility for high volume keywords the estimates provided by Alexa and others do not sync with the volumes expected by achieving rankings for such terms.

      When you begin to look at other tools WordTracker and Keyword Discovery they too indicate some high volumes for keyword searches using different data than Google but also different time periods. It is certainly a maze which needs to be carefully navigated.

      Pete

      • Thanks for the great feedback, Peter!

        You certainly are not alone in being dubious of the stats provided by Google. While doing the research for this post, I often wondered whether the ambiguous initial explanation for the difference between the SKTool and GAKT search volumes was intentional, for obvious reasons.

        Having said that, Google is still the 800lb gorilla in the search space, and can throw more money, more infrastructure, more people, and more brains into making search marketing work better for everyone. They continue to accumulate massive search data that no one else can get their hands on (tools like Wordtracker, Keyword Discovery, Alexa etc can’t even get close). And, perhaps naively, I still believe Google is on average a benevolent giant.

        And you are right on target in suggesting that care and knowledge should rule the day. Hallmarks of a true SEO pro!

    4. Hi Axel

      From Google’s statement that:

      “The updated Keyword Tool combines search data across Google.com and all affiliated search properties. Data from the Search-based Keyword Tool is based solely on searches on Google.com”

      I would understand them to mean that the higher numbers include region specific Google sites (such as http://www.google.co.za) as well as sites where the Google search technology powers their onsite search.

      I would not understand this to mean that the higher number results include clicks on PPC adverts – well not from that statement at any rate.

      Am I missing something?

      Regards
      Liam

      • Great question, Liam! You are really getting to the heart of the matter.

        Even if you interpret Google’s statement literally, then “affiliated search properties” includes their whole search network, including myspace.com, nytimes.com, business.com, ebay.com etc (to my knowledge, the complete Google’s search network is not published anywhere). These properties are not available for organic search listings, only PPC listings. So why has the SEO world been using these numbers to justify SEO market opportunity?

        But, more importantly, I am not sure if Google’s statement should be interpreted so literally. For example:
        1. The Adwords Keyword Tool is the primary tool in Adwords to determine “market size” for PPC ads. So it seems logical that the Adwords Keyword tool covers both the search network and the content network opportunity.
        2. From Google’s perspective, isn’t an ad displayed on an affiliated content site — keyword matched — a form of search-matching?

        Finally, I had a recent conversation with a Google Adwords expert (on the Google payroll) who pretty well confirmed that the Adwords Keyword Volume estimates include keyword-matched PPC activity.

    5. …”These properties are not available for organic search listings, only PPC listings. So why has the SEO world been using these numbers to justify SEO market opportunity?”…

      When I did a search on business.com the listing had a featured list and a sponsored listing. And then would change depending on what you searched for. Sometimes I did get a web listing.

      myspace.com seems to just list web listings.
      nytimes just seems to list its own articles or other news blogs.

      Am I doing something wrong here?

      This is a great post. One of the best I have seen in a long time. Great work.

      • Excellent comment, Ron!

        To best answer your question, let’s take a closer look at business.com with an example. Let’s search for “marketing automation”.

        The search for “marketing automation” returns 6 “featured listings”, 12 “sponsored links”, and 10 “listings”.

        “Sponsored Links” are PPC ads. Pure and simple.
        “Featured Listings” are listings from their PAID directory (business directory advertising)
        “Listings” are also labeled “paid”, with reference to PAID directory.

        By conducting this search I have just increased the July search volume for “marketing automation” by one. And yet there was never any opportunity for my search-engine-optimized, organic page to appear.

    6. Im a bit dissaspointed now because i had previously estimated from google what i thought to be my potential traffic

      • I hear you Chris. Switch to using the search based keyword tool. Much more realistic. You and your clients won’t be disappointed by forecasting unrealistic organic traffic that just doesn’t materialize. And as the old saying goes, 1 hen in the hand is worth 3 in the bush.

    7. I found your blog post because the main keywords for my industrial website are “Environmental Equipment”. Imagine my surprise to read that Google’s keyword tool is that far off! I quit advertising on Google’s Content Network long ago because those clicks do not convert to sales. I had a lot of clicks with no ROI. Thanks so very much for the very relevant information. I only wish I had seen it sooner!

    8. Denise; Thanks for reading and commenting!
      It may be that Google is less far off than you think, because we do work for a company which also has the keywords “Environmental Equipment”. There are some links from their site to ours – so maybe – just maybe – there is a reason for you arriving on our site the way you did.

      Having said that, however, I’m glad you read the post as it does highlight some important considerations in terms of search volumes for organic results versus PPC.
      Good luck with it all.

    9. I agree that you should divide the results given by Google’s Keyword Tool by a factor of about 100. Even though Google says that the Global monthly volume is the “approximate 12-month average of user queries for the keyword on Google search”, it is actually consistently 100 times bigger than the figures given in Google Analytics and the analytics given by my site host.

      The best thing to do is basically ignore the SEO hype that you read about keyword volumes and use raw data from your own site host and, with caution, Google Analytics. Even Google Analytics misses about 70% of my unique visitors!

      I still find it amazing that the world is full of self-proclaimed SEO experts when Google haven’t released any details of their search algorithms for around a decade now. So, similar to religions, the SEO experts can say anything to anyone and no one can prove or disprove what they say. Only the guys at Google can have a real chuckle at some of the dross that you read about the importance of character spaces in meta tags, inbound link page ranks, page descriptions,… It’s the modern day equivalent of the Fairy Tales by the Brothers Grimm.

    10. Hi Russell;
      Thanks for reading and commenting – we do appreciate it.
      While you are, I believe, correct in much of what you say, we are one of those SEO Experts you refer to. And we do help people achieve higher Search Engine Result Page rankings (SERPs). We don’t just make this claim: we measure our performance and offer money-back guarantees to back it up.

      Part of the miss-communication here, I think, is that you are doing this for yourself with an existing site, and can therefore do what you suggest in terms of using “live” data to compare to “guesstimates”. Many people are not in the same position. They are building their website, or renovating it to achieve better results. And its these people who need help often because they don’t know how to establish the right keyword strategy. Knowing that their entire website must be built on the right words and phrases, they hire expert help to ensure that they don’t waste time and money.

      And to address your point about Google not releasing much about algorithms in a while: Google releases information on how to get better results regularly on its Webmaster Tools and its blog. You will find a great deal of helpful information here, and the company SEOMoz does some excellent research as well – their blog, and especially their “White Board Fridays” series of videos is extremely informative. I strongly suggest you check these sources out as they do help one understand the search engines latest approaches.

    11. Very energetic post, I enjoyed that bit. Will there be a part 2?

    12. Hi Genevieve. Thanks for commenting – glad you enjoyed the post.

    13. Fascinating blog! Is your theme custom made or did you download it from somewhere?

      A design like yours with a few simple adjustements would
      really make my blog stand out. Please let me know where you got your theme.
      Bless you

    14. Your style is really unique compared to other people I’ve read stuff from. Many thanks for posting when you have the opportunity, Guess I will just book mark this blog.

    15. [...] they’re looking at here. As helpful as Google AdWords Keyword Tool is (I’m a big user myself), you can’t take the numbers at face value. Those are estimates of ad impressions, not search query volume. If you’re going to use it, make [...]

    Add your comment

    Your email is never published or shared

    Top B2B Blog List

    B2B Marketing


    This website is licensed under the Creative Commons Attribution-Share Alike 2.5 Canada License. To view a copy of this licence, visit http://creativecommons.org/licenses/by-sa/2.5/ca/ or send a letter to Creative Commons, 171 Second Street, Suite 300, San Francisco, California 94105, USA.

    Gossamar is proudly powered by WordPress
    Entries (RSS) and Comments (RSS). 91 queries. 0.265 seconds.