Google Analytics: The Most Instantly-Useful Change of the Last 6 Months

If you use Google Analytics, and your site gets any traffic from search engines, you will love the latest feature they've just added, which was announced on the Webmaster Tools Blog on 4th October.

Here are the stats you used to get when looking at a 'Search Engine Keywords' report:

  • keyword
  • number of visits
  • pages per visit
  • average time on site
  • % new visits
  • bounce rate

So, you could see that you had (say) 50,000 visits for the word 'laptops', but you were left asking questions like "so what?", "is that good, or bad?", and left looking at other tools to try and answer those questions.

So What's New?

Now, here is one of the new 'Search Engine Optimisation: Queries' reports:

I've highlighted the exciting bits there:

  • The red circles are new pieces of data they've added (from 'Webmaster Tools')
  • The pink arrows point to a single example of an instantly actionable piece of information

As you can see, rather than asking "so what?", we can now say "oh, I'm already in position 1 for laptops - no more work needed there... but LOOK - I'm way down in position *nine* for 'cheap laptops', and I'm only getting 250 of a potential 27,000 visits! Let's put some work into improving that."

It's a struggle to think of a business that wouldn't benefit from having instant access to this kind of information.

How to Get at the Information:

To use this, firstly, you must have both 'Google Webmaster Tools' access AND 'Google Analytics Admin' access for the site you'd like to look at.

(if you find yourself struggling with either of those, feel free to pester me on Twitter & I will happily point you in the right direction).

Following that:

  1. Log into Google analytics.
  2. If you see the phrase 'New Version' toward the top-right of the screen, click on that.
  3. Once in the 'new version', select the site you want to look at.
  4. Now, you should already be in the 'My Site' tab (toward the top-left of the screen). If not, click into it.
  5. Choose 'Traffic Sources' from the left-hand navigation.
  6. Within that, click on 'Search Engine Optimisation' & then 'Queries'.
  7. You will see a report similar to the screenshot above.
  8. Weep with joy at the new, simple information that helps you with your job. (ok, maybe that's just me)

Other Extras:

Aside from the 'SEO Queries' report, there are a couple of other useful reports available alongside this:

  • You'll see there is a similar 'Landing Pages' report. Rather than telling you which keywords you can improve on, this shows you which landing pages have potential for improvement.
  • And you'll see there's a 'Geographical Summary' report. This shows you, by country, how many Google Search impressions you get, how many clicks, and what your CTR is. Great for multinational sites.

As with every stat in Google Analytics (perhaps moreso in this case), these numbers will not be flawlessly accurate, but they hand you the basic information - all in one place - to be able to go and improve your site today, tomorrow, and next week. In my opinion these are among the best changes we've seen this year - very small tweaks that make the tool far more useful for almost all websites.

Do drop a note in the comments if you have any further thoughts on this, and do share this post if you feel others would benefit.

Share your thoughts

  • Jeanette commented on February 19, 2012

    Thanks Dan for this post! What makes me think though is the coherence of data from different sources (see e.g. your great post “Using Google Analytics to Audit and Improve SEO”).
    How should one deal with the fact that the number of impressions for a given query (in the Search Engine Optimisation Report) is higher than the number of this query in Google Keyword Tool?

  • Hi James, cheers, you asked about whether the search data may be included in this report. Google could include it since it’s not personally identifiable information and there’s no way for anyone to relate to individuals. Good point about conversion – that’s important of course. Don’t know whether it’s possible to get at that using custom reports or Advance segments – filter visits with sales. It’s going to be clunky though.

    We’ll have to watch for disclosures on this. I was reading some posts last night that show many sites are in the 10-30% range of “not reported”.


  • Hey Dan, Dave

    Sorry the word ‘negate’ wasn’t the best to use. What I meant was reduce the benefit of.

    I like your optimism and I really hope Google doesn’t kill effective keyword analysis for SEO. I know it’s just for logged-in users but with Google + the number of people always logged-in is on the increase. And they will inevitably roll this out from .com across other domains, so the volume of keyword data for reports like the above will diminish over time, at least as I understand it.

    Dave – re your comments, do you mean that the referrer string won’t affect this report? If that’s the case it’s strange that Google would block it on the one hand under the guise of privacy protection and then reveal it in the other hand. I find that rather contradictory. My interpretation is that the keyword data will not be passed at all if you are logged-in, so this will be lost from all reports within GA.

    And even if the new SEO report is still populated with keyword level data, it doesn’t have the engagement and conversion metrics that the traffic sources report does, so you’ll only be getting a small fragment of the overall keyword performance picture. I personally think it’s a sad day for analytics and there is a bigger commercial picture somewhere that Google thinks this will contribute to.

    What do you think gents?


    ps Dan – don’t you just love it when you get excited about something only for the party to be crashed a few weeks later! Keep the articles coming

  • hi, Dave,

    1. My guess is google are still recording all of the info, but will not report it in Google Analytics. (fingers crossed I’m wrong!). With all of the ‘anti competition’ stuff in progress around them favouring their own products, I think they’d be silly to do that. (esp as IBM own Coremetrics & Adobe own Omniture).

    2. I’m not sure rank checkers would be any more useful than this tool. But, I think you’re right – I’m sure their HTTPS stuff will damage the adoption of this, just as it hampers a lot of other analytics stuff. Very sad!

    What does anyone else think?


  • Hello James, Dan + others reading this!

    I wondered who would ask this question – nice one James.

    I agree with Dan’s sentiment, plus another couple of thoughts:

    1. Although searches from logged-in users are not being sent in the referrer string, for standard analytics, Google could still record and report them – here – I haven’t seen anything contra this. Have you?

    2. Maybe if this tool becomes less useful, marketers will revert more to good old rank checkers which Google has been trying to stop adoption of years – hence the creation of this tool?

  • Hey Dan,

    Just catching up on reading and found this – spot on post but don’t you think that the new secure search announcement effectively negates this report as Google said that the keyword from the search won’t be passed through? I know this is currently only on .com and for logged-in Google account users, but it’s likely more and more people will browse when logged-in and Google will inevitably roll out secure to other domains.



    • hiya, James, how are you?

      A lot happens in a month, eh? Google’s ‘secure search’ announcement came after this post. I very much dislike the secure search announcement, as you’ll know from Twitter, the blog post over on econsultancy, etc.

      I don’t think it totally negates the usefulness of this feature, but it certainly messes it up a bit. Here’s an optimistic thought:

      Google will hide any search terms where visitors are logged in. They replace those with “(not provided)” instead. In theory, those should affect every search term to an equal extent (in practice not true, but let’s ignore that). Therefore, the questions this report allows you to answer will be equally valuable.

      Those questions are: “where am I currently?” & “where do I have the potential to improve?”. These reports will still answer both of those questions for any keyword with decent volume.



  • Jeanette commented on October 28, 2011

    Thanks for another great article, Dan!

    Although these questions are relating more to GWT than to GA, they are always coming up:

    – How can it be that the number of impressions in this report (such in GWT) is much higher than the search volume of the same term in Google Keyword Tool (using exact match)?

    – How is the number of clicks in this report relating to the number of visits in sources -> organic report? In all reports I saw, the latter number is higher.

    Your ideas?

  • Dan

    As always many thanks for highlighting this to us and for demonstrating how useful this will be. Looks like i am in for a long night of analysis but really looking forward to being able to take some real positive actions out of it.


  • It’s a great new feature, though from experience I’ve had with it so far their position reporting is quite off! A term we have ranked for consistently at position 3 for 8 weeks+ is reported as being at position 12…

    • Hmm, that’s not great is it Adam? I’m seeing similar magnitude of inaccuracy for some our terms even when you take personalisation and country into account.

      This charge was levelled at the original GWT version too. I wonder whether we will see any systematic comparisons. You would think Google would get SERPs position accurate…

      Is anyone else seeing inaccuracy on position reporting?

      • I thought this was a great point, so I quickly ran an analysis for one of the companies I work for. Their site gets >300,000 visits a month from search, so there should be a decent spread.

        From the top 100 search terms (according to the SEO Queries report)
        In 14 cases, the report exactly matched up with the ranking I see in Google.
        In 38 cases, it was within 1 point of actual rank.
        In 22 cases, it was within 2 points of actual rank.

        So, from that very small sample, there was a 74% chance that the report will be ‘within 2 points’ of the actual ranking number.

        The other interesting thing to remember here is: This is giving you the ranking across all countries. My guess is, the more countries you appear in, the more your ranking will fluctuate. (outside of the effects of personalisation & +1 metrics.

        Any thoughts?

        • Richard Fergie commented on October 5, 2011

          Try tracking rankings using the tracking number passed in the ref string for some queries. You’ll see that rankings can vary quite widely for a given query so taking an average hides a lot of what is actually going on.

          Geo seemed to be a factor, but sometimes I think Google just displays weird rankings.

  • Hi,

    Thanks for the great article. This was very useful indeed. I think the new Google Analytics in general has really improved (from a beginners’ point of view anyway) and being able to view the ranking of your content is excellent. I definitely want to pay more attention to this after the Christmas rush!

    I appreciate this is unrelated to this thread but on a personal note to Dave Chaffey – I used your Internet Marketing textbook for my degree – I still have it at home and think it is a great book. It must have taken you ages!

    I also agree that it is great seeing yourself and the CEO of Econsultancy comment on here too. Great interaction from two digital marketing heavyweights and shows us all how we need to keep learning in this ever-changing and complex industry.

    Many thanks,

    Gareth Powell

    The Emporium Direct

    • Hi Gareth,

      thanks for taking the time out to credit my book – I’m currently updating the 5th edition, so it’s always nice to hear from folks who have found it useful, since updating is a BIG task now shared with Fiona Ellis-Chadwick.

  • Thanks for this Dan.

    The one final piece of information on the SEO report is ‘Goals’. It would be great if you could see the query report with ‘conversion rate’ at the end without having to flip between reports and filtering by keyword/phrase.

    It would be great to say ‘keyword1’ had ‘x number of impressions’ and achieves a ‘click through rate of x%’ and also has a ‘conversion rate of x%’ based on an average position of 4.5.

    And then in 6 months time, you may be able to say that the conversion rate decreased by 5% due to the drop in impressions or position or CTR.

    More to come maybe??

  • This is excellent news – it’s something I’ve been waiting for since GWT and Analytics first linked up, but sounds even more useful than I hoped.

    I’m just off to play with the data now..

  • Thanks guys. Lots of interesting stuff in here. I’ve had a quick look through for and two things certainly are apparent:
    1. We’re getting a lot of clicks from SEO to pages that we wouldn’t have thought ‘important’ (e.g. press releases) which further reinforces the message about not obsessing with your homepage etc.
    2. It’s interesting for us, as we’re expanding in the US, to see that the top landing page for US traffic *isn’t* the homepage (as it is for the UK) which reinforces to us how relatively unknown our brand is in the US compared to UK.

    • Cheers Ashley, good to see a CEO/director actually using their analytics to make improvements!

      What I’ve noticed is that most blog content is ephemeral in terms of driving traffic (as is well known) like this example below, I was discussing with a commenter here”

      Although new blog posts can contribute a fair amount of traffic, the trick from an SEO POV seems to be to find how to generate sustained traffic from SEO for pages which also generate leads/sales. I’m finding this new tool useful for identifying and improving these pages which visitors reach when they have a specific need.

      Hi Katie, let us know what you find!

      • “good to see a CEO/director actually using their analytics to make improvements!” – I’d very much hope this will become the norm. Any CEO that isn’t using data in a sensible way (not just web data) to improve the business shouldn’t be in that job?!

        …”the trick from an SEO POV seems to be to find how to generate sustained traffic from SEO for pages which also generate leads/sales.” Absolutely, that is really the much more important challenge. I’ve got a lot I could say on that! But broadly I do believe the ‘domain authority’ thinking that if you build SEO value to, say, the blog section of a site which is actually selling, say, reports or training (like us), then it does have a halo effect on the whole domain. That said the reality is that most of the *traffic* comes to the free blog part of the site and not the part which makes money. Lots of experiments we have planned around that. Most of those we’ve tried so far haven’t worked very well to be honest. But at least we know that!

  • Thanks for alerting us to this great new feature Dan! I get the feed from Google Webmaster Tools, but didn’t realise this update allows you to see the data from within Google Analytics with the context of actual rather than estimated clicks (there was much discussion about the accuracy of this feature originally).

    I have now set this up for Smart Insights and I can see that the big new feature here, previously unavailable for SEOs I think is the Landing Page reporting tool.

    This shows impressions too and average position across a range of keyphrases, so we now have “landing page potential” to look at and try and improve. In reality this doesn’t work so well since most of the average positions are quite high – mine are in the in range 10+ even for top landing pages because it must be taken across a large number of keyphrases. It would be useful to know know in more detail how they calculate this.

    One other observation for folks looking at this tool is that naturally ranking will vary by country and if you add a secondary dimension of country you can see this will how position and CTR varies. This is confusing with all the countries though, so it’s best to apply an advanced filter (or use Adv.segmentation) of “United Kingdom” or country of interest.

Get FREE marketing planning templates

Start your Digital Marketing Plan today with our free Basic membership.

  • FREE fast start guides to review your approach
  • FREE digital marketing plan templates
  • FREE alerts on the latest developments

Need help with your Google Analytics?

Get more from your digital marketing with in-company or remote training and consultancy from Smart Insights

Get FREE marketing planning templates

Start your Digital Marketing Plan today with our free Basic membership.

  • FREE fast start guides to review your approach
  • FREE digital marketing plan templates
  • FREE alerts on the latest developments