How does CTR relate to SERPS Position?

[Zippy-Bungle]

I'd always though that CTR would increase with SERPS position but my data doesn't really seem to show that.

I used google analytics/acquisition/SEO/Queries  and looked at the top 1800 queries. Then I filtered those with a CTR of 0 and for branded keywords.

The result was as shown - a big glob with no real pattern other than a long tail of keywords with a 100% CTR

Does this tell me anything useful or am I just wasting my time?

Its probably worth stating that only 12 of my keywords had over 1000 impressions and none over 10,000 (all in top 10 though) - but thats a function of my industry sector

gJgSNGL.png

[CommT]

Hi Zippy-Bungle,

If your rankings improve then your CTR is likely to improve too. To affect CTR you can use the meta description and title tags - the more compelling your meta des the more likely it'll get clicks. I sometimes use high performing ad text for this if it's relevant.

As to whether an increase in CTR would relate to an increase in rankings - I don't really know. However, I did notice that our CTR dropped slightly and then our listings dropped a little too - whether this is a coincidence or not I couldn't tell you.

I think it would be very difficult for Google to use CTR, time on site and bounce as ranking signals. For example, if someone just wants my company's phone number, they may do a brand name search, find the number at the top of our pages and call us but analytics/google  would count this as a bounce! So, we've been 'useful' but we've not encouraged the visitor to click into deeper pages (they found what they were looking for on the first page they visited) - so how does google know when a page has been 'useful' or not? (obviously I know links play a role here, I'm just speaking from a data perspective).

So, I suspect google doesn't use these signals, because it seems pretty darn difficult to determine what is a real bounce and what isn't (as in a real bounce would be where the site you click on is useless for the query). As for time on site... well, if you follow google's instructions and make your site faster (we halved our site speed in April) then time on site will inevitably reduce (ours did!) so, again I don't know how google would use this (unless it was in conjunction with site speed). As for CTR - that only tells one half of the story, so again, not sure how google could use it. I know they use it for adwords, but it's in their best interests for the most likely to be clicked ad to be highly prominent and any reduction in cpc as a result is easily mitigated by the fact that it gets more clicks (stack 'em high, sell 'em cheap)

However, I do think google probably uses a metric similar to bounce, when users use the back button to immediately leave a page they've just landed on and end up back on the same serp as they started from. This is just my theory, and the reason I think they use something like this is because if I was them I would. I will probably get shouted down here by everyone else, and this is just my opinion so do with it what you will!

[Zippy-Bungle]

Hi Tom, yes that's what I meant,. Never thought about it the other way round  so I'm glad for the confusion as I learned something that I hadn't expected to learn.

Ive ve still got the issue with the data blob though not seeming to correspond  with what other sources show. Is it a function of low search volumes or something else?

[TomRayner]

Beg your pardon, I think I misinterpreted your initial question.  My answer addressed "Would an increase in CTR result in an increase in rankings" (No), but I think you meant "Would an increase in rankings result in an increase in CTR" (Usually yes)

Sorry about that, was that what you meant?

[Zippy-Bungle]

Thanks. That's good to know. I thought there was a case a few years ago where some Yahoo data got into the open and it showed that there was a correlation.

Checking on that I discovered this Moz article (which is more recent) which seems to imply that there is a relation - or have I misunderstood your answer, the article or both?

[TomRayner]

Your data shows you what is generally accepted - there is no relation between CTR and organic search.

Certainly with what Google tells us it has no baring at all - although we should always take what they say with a pinch of salt, my own tests and most other people I know agree with this.  Some people think the time on site/bounce rate may have an affect, but again I don't think this is the case either.

The only time where CTR comes into play is for PPC, where it can affect your cost-per-lick and quality score.

For SEO, it's widely agreed that there is no relation.

Hope this helps.