SERP Ajaxification Gives Your GA Data Juice!
I read a great post yesterday on Joost’s blog by Andre Scholten that discussed tracking keyword ranking positions in Google Analytics – genius and well, I could babble on about it forever as it has been the missing part in any analysis for like, ever! Excellent stuff guys!
It’s all been made possible by the ajaxification of the Google serps . Matt Cutts talked about this ajaxification back in March, (you can see it on youtube here ) saying that Google were testing and that < 1% of users would see this behaviour. Based on my numbers, it would seem that this test has been ramped up some as at least 8% of my visitors used this yesterday to reach the site I’m discussing.
So, with all this in mind I thought it would be cool to set it up and to see what kinds of traffic share each position gave. It’s a limited sample of 1 days data( follow up post will follow with bigger numbers) showing amongst other things the importance of position one in Goog for ones keywords and the value of choosing a good SEO to get you there.
It’s also really cool to see what types of keywords, en masse are appearing and providing traffic for the various positions, especially for the longer tail stuff. The patterns and query types also help inform subsequent content production choices as well as any on page tweaks I might decide to employ subsequently.
A few words on factors that influence CTR first though
Now, granted, it’s not a perfect representation as there are many factors that influence CTR. I can hear some readers saying hang on a minute, 75%, so what, it’s a limited sample, on just one vertical that may well be devoid of all manner of other signals that influence such things. Yep – I agree too, it is but a snapshot, it is but a sample, but it’s an insight nonetheless. I won’t reveal the site or the niche, but I will reveal that it’s in travel. I accept completely too, that no vertical is the same, all have their nuances – the nature of the query, its length, user motive, all affect the decision of the user in their click journey.
For those who don’t know, CTR ( click through rate) is influenced by all sorts of things like; calls to action in the snippet displayed by Google, instances of keywords in the title, url or url file path that match the query.
(The image below shows a typical example for the query’ play poker’ showing the instance of the keywords that match the query)
And then there’s other factors like ‘site links’, ‘indentation’, ‘double facing ppc and organic’, ‘brand awareness’, ‘universal/blended search’, ‘local search’, ‘news results’, ‘PPC ad position’ to name but a few.
Content and Site Type Traffic Considerations
As for traffic itself, again you might have a site that is huge and is accessed by millions of long tail queries that account for the majority of your traffic. OTOH, you might have a site that is just 4 pages deep but carries the best review of a product ever, where one keyword like ‘product review’ accounts for the majority of your traffic. IOW, no two domains are ever the the same – each site is different and is subject to content, depth, breadth, authority, age, location etc.
So having said all that I’m almost inclined to say, what’s the point in even discussing it if every query is different, if its dependant upon so many variables what is the point even! But I won’t, as maybe it’ll encourage others to do the same and share their findings. The more we share, the more we learn, the more we can best serve ourselves and our clients, right?
Lets not confuse things either…
First off, my data from GA simply reveals that 75% of sampled visitors came to the site through clicking on an organic position one result. It does not mean that in my entire niche that 75% of people always click on the position one result. Neither does it mean that in every niche or vertical that 75% of people click position one either, as we touched upon previously, it’s complicated and subject to widely varying factors.
No discussion on CTR’s generally, would be complete without at least touching upon previously known related stuff, so it’s most certainly worth talking about the AOL data set from back in the day that revealed interesting findings in that it showed that aggregately, over a few million queries, 40% of people clicked through to the position one result – if we analysed that data further, we’d see all sorts of contradictions and different outcomes as we pared and refined the data set.
If we were able to see a similar result set today, then there is no doubt that universal search, news, images etc will have t impacted such a figure and for some organic queries today, you’d expect that figure to be a whole lot lower or higher even.Two competing theories being that one; people would be dissatisfied with the 1 or 2 organic results followed by news and local or images so would scroll down further in an effort to get more choice, and two; people are lazy and are therefore more inclined to glaze over the rich media or news or local results preferring to click the organic results at position 1, therefore increasing the CTR and value of #1.I couldn’t possibly do that whole debate justice here, as it’s a massive multi facted issue full of all sorts of fascinating twists and turns which I don’t have the time to get into to, but it’s a debate to bear in mind when you look at this stuff.
Anyhow, without further ado here is the image showing the CTRs from Organic Google SERPs for 582 visitors yesterday who used ajax Google SERPs to arrive at the domain, revealing that 75.4% of traffic yesterday came from position #1 results.
It’s interesting to note too that position 10 got a greater percentage of clicks than position 8.
If you have a client who uses GA, then do yourself, and them a huge favour and implement what Joost and Andre have shown, do it now!