The impressions and clicks don't seem to correlate on Google Webmaster Tools for my site. Whereas the impressions vary widely, the clicks stay constant. I would expect the clicks to my site to go up (even if slightly) when the impressions go up.

Sample days: 2,522 impressions, 91 clicks (3.6% CTR). 1,038 impressions, 54 clicks (5.2% CTR). 2,472 impressions, 104 clicks (4.2% CTR). 1,123 impressions, 70 clicks (6.2% CTR).

So whenever the impressions get less, the CTR rises a lot, when impressions get more, the CTR drops. I have only given an example of four days but this behavior is consistent throughout the 90 days that GWT shows.

It will probably be tempting to put this down to chance, but looking at the graph on GWT it seems not to be a matter of chance. The impressions line varies between 1000 and 2500, with huge jumps up and down, whereas the clicks line stays completely flat, around 70-100.

Why is that?

  • 1
    BTW- I like the questions you are asking on this site. They are good questions and well formulated. Thanks! – closetnoc Dec 29 '14 at 16:52

You actually have good click through rates (CTR)! Impressions can be more than one per page and people often confuse these numbers when looking at them. I look at Queries vs. Clicks. If your ratio is something like 1:1 (1 to 1) or approaching it, then you are doing rather well. A click through rate of 1% can be considered good by some. More than that, then you are doing really well. Of course, some sites will enjoy a very high CTR. What I do not like is this page by Google. In fact, quite a few Google metrics are misleading and/or does not tell me what I really need to know. This is one of those pages that can be improved somewhat and people get confused over.

  • 1
    I have almost 600,000 pages and only about 200 users per day. (sad ain't it?) It is all relative. I have noticed that impressions will go up and down based upon daily trends and clicks often hover within a range. I think that much of this is Google returning results that are not exactly what the user is looking for. Remember that %15 of all searches each day have never been seen before. Google tries to match pages to the search query and uses CTR and bounce rates to tune it's SERPs. Also, some sites will do well over other sites because of what they are. Some are in competitive spaces. – closetnoc Dec 29 '14 at 16:24
  • 1
    Ok thank you. I used to have more visitors when I had only 50 pages, working on the site part-time, then I stopped writing for more than half a year, and the visitor numbers dropped. Now I have been working on the site for 5.5 months full-time and am not seeing a growth in traffic at all, and not reaching previous figures. That's why I am thinking that something is wrong but sounds like it is not the impressions/ctr ratio I have to worry about. – CHR Dec 29 '14 at 16:33
  • 1
    I am in a highly competitive market and do not expect much. I have to compete with sites within two markets because of keyword overlap. What I concern myself with is trending. I do not have a high number of users, but new vs. returning user numbers is what I concern myself with most. The rest is highly volatile due to trending. My CTR is 1% which is good for my market. For me, page count is the most important thing since CTR is often very low for the types of searches that return users to my site. If my CTR remains steady and the returning user count increases, I will do well over time. – closetnoc Dec 29 '14 at 16:35
  • 2
    That is a long time to go stagnant. Google and Bing like freshness and that is a huge metric for SERP placement and takes a long time to turn around. You are just about to see up-turns I am sure. 5.5 months is not enough time to significantly change the effect of 1.5 years of being stagnant. Look for trending upwards even if it is small. I let my site sit for 10 years! It is taking a year or so to start turning the trend back upward. I suspect you will see results again soon enough. Persistence pays off. Keep the faith. It will work out! I promise. – closetnoc Dec 29 '14 at 16:39
  • 1
    All of that stuff is important too, but nothing replaces content and creating content steadily over time. Consistency is important. In between, you can do the other stuff. I will admit my site sucks! I am working on new static content, images, cleaner template and HTML, and so on, but not that the price of content which I do semi-automatically with some batch processes. BTW- I am not sure breadcrumbs are the be-all end-all that people say they are. They are good for blogs when they help the user, but I am removing them from my site. I see no advantage for my site. – closetnoc Dec 29 '14 at 17:03

Click through rate is not a constant but depends on:

  • The ranking -- The result in the first position may have a 25% CTR, second a 10% CTR, down to tenth with a 1% CTR
  • How the listing looks -- Does the result match the query well? Are the keywords bold? Is there a call to action?

You don't say what Google Webmaster Tools reports as your average position, but my guess is that the ones with the better CTR are ranking better.

  • I'm not sure I understand this answer. The CTR is of course not constant but especially then you would expect a certain amount of variation. But the CTR stays the same, where impressions vary a lot. It's as if some of the impressions are 'fake' impressions and not really shown in search results. – CHR Dec 29 '14 at 20:13
  • Some queries have more impressions than others. You end up being able to rank better for less popular queries. I don't see any fake impressions from the data you shared in your question. – Stephen Ostermiller Dec 29 '14 at 20:15
  • Yes but I am looking at the whole site not specific queries... and see a strange disconnect between amount of impressions and the clicks. Sounds like you are saying that it is a coincidence that I always get a much better CTR when there are less impressions and vice versa. – CHR Dec 29 '14 at 20:26
  • 1
    Not coincidence. You rank higher when there are fewer impressions because there is less competition. – Stephen Ostermiller Dec 29 '14 at 20:27
  • 1
    Yes, it is typical. – Stephen Ostermiller Dec 29 '14 at 20:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.