24 Jul 2017

Announcing: 2017 Google Search Click Through Rate Study

Google Click Through StudyIMN is pleased to announce our Summer 2017 Click Through Rate Study, now available in Whitepaper format in the Resources section of our website!

It’s been a while since we’ve last written about this topic, and since last summer, there have been multiple changes to both the algorithms that determine the organic rankings AND the visual presentation of the SERPs.

These changes include:

  • Mobile Friendly Update (#2!)
  • 4 Pack of Ads
  • Penguin 4.0
  • Interstitial Penalty
  • Fred and other un-named or un-confirmed core algorithmic updates

Because of these changes, along with all the other changes seen in the last 12 months, IMN wanted to know: What’s the actual click through rate for a #1 ranking look like these days? How about a #10? A Page 2 Ranking? In that spirit, IMN conduced our own internal Click-Through Rate (CTR) study for Summer 2017.

A more detailed breakdown of our methodology, including data sources, sizes, date ranges, and cohort segmentation can be found in the actual Whitepaper, but to summarize, we used 90 days’ worth of data from Google Search Console from a well-curated list of our own clients.

We analyzed 20,000 queries, which in a 3 month period saw over 64 Million Impressions and saw over 4 Million clicks. The sites were fairly evenly divided between Business to Consumer (B2C) and Business to Business (B2B) websites. IMN also distinguished between Branded Queries and Non-Branded Queries as well.

So, what did we find?

Overall, the CTR for a #1 ranking for All Queries in our study was just over 21%. This is lower than reported in prior CTR studies released in the past, including the AWR Study from 2014, the Chitika study from 2013, and others. Data sets, sizes and methodologies differ greatly for CTR studies, so we invite you to compare these methodologies and how these might impact results.

A #2 ranking secured just over 10% of the clicks and a #3 ranking was the last position to be above 5%, at just above a 7.5% CTR. Rankings at Positions 4-8 were in the 5% to 3% range, and then CTRs leveled off for the most part.

Surprisingly (at least to us!), the CTRs for All Queries did not drop below 1% for any position within the Top 20 spots of Google, and there is a slight dip UPWARDS in CTR on the top half of Page 2 compared to the bottom of Page 1. The changes to the algorithms and SERPs may not have dramatically altered total Impressions on Page 2, BUT there were definitely more Page 2 Clicks, in aggregate, than we expected.

Another interesting trend we found was just how many Impressions for Queries on Pages 1 and 2 of Google result in No Clicks at all – between 25% and 30% of them! With the introduction of the Knowledge Box, the other Rich Card formats, all of the various carousels and blended universal result types, and of course the expanded 4 pack of Paid Ads, this isn’t too surprising to us, but it’s good to see our suspicions confirmed – between 1 out of 3 and 1 out of 4 Google searches end with No Clicks!

IMN also differentiated between rankings for B2B and B2C sites:

The overall CTRs for them were fairly comparable, but the main differences were that B2B sites saw higher CTRs on the top half of Page 1, but slightly lower CTRs on Page 2, in aggregate.

And finally, IMN also distinguished between Branded and Non-Branded queries:

Our main findings were that Branded Queries see MUCH higher CTRs on the top half of Page 1, but that 99% of all Branded Query clicks came from the Top 3 spots.

Frankly, our data set size for Branded queries past Position 5 was quite limited, but we suspect that despite whatever visibility a Branded Query might see past Position 5, it’s not likely to see many clicks.

Our highest-level findings for the Summer 2017 CTR Study include:

  • On average, a #1 ranking in the SERPs had a CTR in the low 20% range. IMN found that less than half (~40%) of the #1 rankings saw a CTR at or above 30%.
Less than half (~40%) of the #1 rankings saw a CTR at or above 30%.Click To Tweet
  • A #2 ranking was usually about half of a what a #1 ranking secured – in the 10% range.
A #2 ranking is usually about half of a what a #1 ranking secures: ~10% CTRClick To Tweet
  • CTRs continued to lower, albeit more slowly, to about the 1% to 2% range by the bottom of Page 1, but Page 2 CTRs were not that far removed from bottom of Page 1 CTRs.
  • 28% of All Queries did not see a Page 1 or Page 2 Click at all.
  • Branded queries saw higher CTRs by far than non-branded queries for the top half of Page 1, but 99% of the clicks went to the first 3 positions for Branded queries. Data size and reliability past Position 5 for Branded queries was limited in our data sets.
Branded queries have higher CTRs than non-branded queriesClick To Tweet
  • B2B sites tended to have higher CTRs on the top half of Page 1 compared to B2C sites, about equal CTRs on the bottom half of Page 1, and then slightly lower CTRs on Page 2 compared to B2C sites.
B2B sites tend to have higher CTRs on the top half of Google's Page 1 compared to B2C sitesClick To Tweet

To see our full CTR Study, click here. Please drop any questions or thoughts in the Comments below and we’ll do our best to answer them!

Comments

  1. DNN July 24, 2017 at 2:53 PM

    A few things I’ve intensely focused on to improve my Google CTR and boost site retention:

    Evergreen content marketing

    Less ads

    More relationships with fellow bloggers so i can build lasting relationships and encourage bloggers to mention my content in future blog posts

    Stop thinking in the moment about making money

    Put people before profits

    Create more YouTube videos with lengthier video descriptions with links in the header, center, and footer pointing to specific pages

    Get in the daily mindset or creating content either everyday or often so I can naturally increase my traffic from Bing, Google, and YaHoO!, all while having my target reading audience share my content and give me free advertising on Pinterest, Twitter, Facebook, LinkedIn, and Google+.

    I’m so glad affiliate marketing and content marketing is here to stay because honestly, I feel I can’t work 35 years like my mother did for the City of New York in Brooklyn “working for the man on a 9 to 5 day job.”

  2. Ash Nallawalla July 24, 2017 at 7:23 PM

    Thanks for sharing this, particularly for all positions up to position #20 and distinguishing between B2B and B2C. Others had been more protective of their findings when I asked.

    1. DNN July 28, 2017 at 10:22 AM

      Ash,

      Just wondering who are the “others’ you mentioned that were protective. You can email me that info if you want to through the site contact us.

  3. Karan Chauhan July 25, 2017 at 1:44 AM

    Is case study all about the nice graphs? Can I see list of queries which you’ve analysis?

    1. Dan O'Leary July 28, 2017 at 4:11 PM

      Hi Karan,

      Sorry, but we cannot share that level of detail externally, in order to protect client confidentiality. We do share some details in the Methodology section of the study about the types of sites we chose, however.

  4. mads July 25, 2017 at 2:30 AM

    Nice post, I have just a few questions I hope you can answer;
    1. How did you collect the date?
    2. How big was the population?
    3. Over how long a timeperiod did you collect the data?
    4. What country(s) did you study?
    5. How did you sellect the population?

    Hope you will answer these questions so I can validate your findings and use your article(link to it).

    1. Dan O'Leary July 28, 2017 at 4:16 PM

      Hi Mads!
      Some of these have been answered in the Case Study, but:
      1) How did you collect the date? – We expanded GSC query data to the maximum 90 day period for the clients. We began starting the study at the very end of Q2 2016, hoping to collect data from that same quarter.
      2. How big was the population? – 20 clients, 20,000 queries, over 64 million impressions and over 4 million clicks.
      3. Over how long a timeperiod did you collect the data? – Over 2-3 days, we gathered the prior 90 days’ worth of GSC query data.
      4. What country(s) did you study? – United State only.
      5. How did you sellect the population? – We sampled within our own client base, with the goal of selecting a wide range of sites from different verticals and with different business models. At the highest level, half of them were Business to Consumer (B2C) and half were Business to Business (B2B).

      1. mads August 11, 2017 at 2:14 AM

        Thanks for the reply Dan!
        Very usefull. When I get the time i would do the same for my clients in Denmark.
        A last question, how long did the survey take you (how many hours)?

  5. Raphaël July 25, 2017 at 4:37 AM

    Thanks a lot for this interesting overview.
    What could be very interesting to analyse too are the CPCs evolution depending on the avg. position.
    Is pos. 2 twice more expensive than position 3 or more? what about getting from 2 to 1?
    And then relate CPC ratio vs. CTR. Of course it would be an avg overview but could be interesting to compare with our own activities.

    1. Raphaël July 25, 2017 at 5:34 AM

      And of course a separate view by device would be very interesting too!

      1. Dan O'Leary July 28, 2017 at 4:23 PM

        Raphael,
        We will be looking to segment out mobile vs desktop in a follow-up or in our next case study. Thanks for the feedback and suggestion!

    2. Dan O'Leary July 28, 2017 at 4:20 PM

      Hi Raphael,
      For simplicity and consistency, we did not (originally) overlay CPC data onto the queries, because that would involve an additional and slightly different data set and source. However, we can tell you in that in an internal follow-up on our study, we’ve found that about 2/3 of the queries we reviewed had reported CPC figures according to SEMRush. 1/3 of the queries had no reported CPCs according to SEMRush. If and when we find any trends regarding CPC and CTR, we will definitely update and share! Thanks for the feedback!!!

  6. Avi B July 25, 2017 at 8:31 AM

    Dan,
    I just love these kids of study. Something that actually gives you something of value.

    What was the median for each position?
    Was there a big variation between KWs click distribution?
    I think a bell curve graph for each position would really help show the full picture on click distribution.

    1. Dan O'Leary July 28, 2017 at 4:34 PM

      Hi Avi B!
      1) We didn’t calculate that initially or share that within the study, but interesting question. We will try and provide a follow-up response as soon as we can!
      2) We’re not clear on what you mean by “variation between KWs click distribution?”, but every position from 1 through 20 had multiple queries that had as low as a 0% CTR and as high as 100% CTR.
      3) A bell curve for each position was something we initially considered, but it would be far too complicated within a single graph and would be too many unique graphs for what we wanted in the study and the blog post. However, we’ll see what we can do in terms of a follow up. Thanks for the suggestions and feedback!

  7. Colin Campbell July 25, 2017 at 10:11 AM

    Nice Article!!

    At Instar Group we focus our efforts to achieve client goals within the rapidly growing digital space, be it within:
    Social Media
    Content Marketing
    Advertising
    Design
    Events
    Social Media Marketing & Content Marketing are two intertwined specialties. We understand both. The Instar Group team prides itself on topic research & personalized campaign design to suite your brand, personality or practice.

  8. Mike July 25, 2017 at 11:07 AM

    This is a much-needed information to tally what is going on in the Digital World. The statistics are a little different than the previous year.

    This also states that making content has no value if it is not ranked in teh top 5 positions on Google.

    1. Dan O'Leary July 28, 2017 at 4:43 PM

      Hi Mike!
      Thanks for reading! While we agree that the CTRs beyond position 5 were eye-opening, we will have to disagree as to whether content ranking past that position has NO value. As mentioned in the study, we found a consistent CTR above 1% on every spot on Page 2, which was something we did not expect initially. Moreover, although there are certainly less organic search clicks per query, that doesn’t necessarily means there’s less value. When the alternative is renting, even when there’s less beach front real estate, that doesn’t diminish its value. In some instances, it may raise it!

      Thanks for reading and for the feedback!

  9. Mayank Rana July 26, 2017 at 12:48 AM

    A great case study. Thanks for sharing this with us. Keep Updating us.

    1. Dan O'Leary July 28, 2017 at 4:47 PM

      Hi Mayank Rana!
      Thank you for reading. We will be looking to conduct similar studies more regularly moving forward, and we will happy to provide updates and follow ups as we can. Thanks for the feedback!

  10. Barry Garner July 26, 2017 at 3:42 AM

    Very valuable info, I would be curious as to what the CTR is for Maps listings. With the new messaging feature in Maps, I see this as the Golden Nugget for local businesses and I see more and more that Google is going the pay 2 play route, Thank you for your insights, I will be giving it a share.

    1. Dan O'Leary July 28, 2017 at 4:38 PM

      Hi Barry Garner,
      For the context of this study, we weren’t able to distinguish to our liking (yet) which queries resulted in maps, carousels or other blended/universal results in the SERPs. However, the study itself was born out of the collective internal hunches that such features were impacting historically reported CTRs. As we refine our methodologies further and we garner any such insights, we will provide updates and follow ups. Thanks for the suggestions and feedback!

Leave a Reply