Conflicting average position data from Google Search Console?
-
I'm looking at Google Search Console data in Google Analytics, specifically Average Position as given in the Landing Page report, and the same metric broken out by mobile and desktop in the Devices report.
In the Landing Page report, I see an aggregated average position that's much higher/worse than an actual average of what is reported for mobile, desktop and tablet traffic under the Device reporting.
For example:
- Mobile: 5
- Desktop: 5
- Tablet: 5
So the average still should be roughly 5, right? Why would the Landing Page then show an aggregate Average Position of 8? I wouldn't expect to see a precisely same average given that different device types have different proportions that could render differently when the buckets are combined, but this is a huge swing. In fact, the aggregate Average Position as given in the top level Devices report is closer to 5 than to the 8 shown in the Landing Pages report. (These aren't actual numbers, but are illustrative of what I'm seeing, by the way.) Unless I'm missing some vital difference in the way that Average Position is reporting for the Landing Page report versus the Device reports, it doesn't seem like this should be possible. What am I missing?
-
Howdy.
So, here are my thoughts and findings.
The discrepancy starts in Google Search itself. So it's either algorithmical/calculational bug or there are other metrics used. And i'm inclined to believe the second version.
Here is my idea - they use sessions or clicks as a metric together with avg position. As far as I know they use this system in other metrics as well. So, here is data for our website, which checks out with all the reports:
I got a page https://www.regexseo.com/luxury-websites-houston.php and the avg position reported is 22. If i look at the breakdown, it's got 96% of traffic going to desktop and 22 avg position, 2% to mobile with 15 avg position and 0.05% tablet with 14 avg position.
So, the report checks out, if you think about it like this:
If to 96% of users your page is ranking #22, and to 2% of users it ranks in #15, wouldn't it make sense to say that on average your page is ranked #22.I guess it's more like a weighed avg position then pure avg position.
Hope it makes sense
Got a burning SEO question?
Subscribe to Moz Pro to gain full access to Q&A, answer questions, and ask your own.
Browse Questions
Explore more categories
-
Moz Tools
Chat with the community about the Moz tools.
-
SEO Tactics
Discuss the SEO process with fellow marketers
-
Community
Discuss industry events, jobs, and news!
-
Digital Marketing
Chat about tactics outside of SEO
-
Research & Trends
Dive into research and trends in the search industry.
-
Support
Connect on product support and feature requests.
Related Questions
-
Can't see backlinks in Search Console
Hello, We have 7 sites and have noticed that for one site, we don't see any backlink info in Search Console even if it's been linked for over 8 months. Other tools show thousands of backlinks, but Search Console is still pending. I also see very little info in regards to Performance even if we are getting thousands of hits a day. Could this be a sign of a bigger problem? In summary, the site is up and running, getting hits, getting backlinks, but Search Console still looks like it's a new site with no activity several months after being installed.
Reporting & Analytics | | CJolicoeur0 -
Query on Google Experiment?
Hello All, I am doing A/B testing via google analytic experiment now my query is for my ecommerce site homepage i am trying to add newsletter in 3 different way so it will be variation A, Variation B and Variation C. So my query is what should be then original page? Currently there is no newsletter form on homepage. Do I consider original page as Variation A? i.e. abcd.com?variation-A as original page? Then google will decide winner between original page i.e. variation A and Variation B & C Thanks!
Reporting & Analytics | | dsouzac0 -
Google Analytics, new property or views?
Hi All We have an existing Google analytics account which was customised relatively heavily a few years back, with custom variables, e-commerce tracking and the a large number of filters. We have determined that the custom variables and e-commerce tracking, at least for now, are no longer required and actually complicating our lives. We want to simplify the account and add a fresh set of goals and events. Our initial train of thought was to create an entire new property with the usual views, raw, test, main etc. We are proposing to retain both properties (6 months) so we can still do some comparative stuff. We're now wondering if we should retain the existing property and just create new views, the downside being that whilst we can compare data with previous periods it's meaningless to some degree as the filters etc are different. The plus of more views is that we're not firing the tag twice retaining two properties. Views and opinions welcome. Thanks
Reporting & Analytics | | mde9110 -
How to devide Google traffic into local google traffic (.com , .uk , .cn etc) in Analytics
Hi, friends In google analytics , we can see search traffic from google, but is it possible to deeply see from which local google? eg: .com , .uk, .hk etc? Thanks alots
Reporting & Analytics | | topchinaseo
Boson0 -
Google WebMasters Tool - Preferred Domain
I just added Google Analytics to my wordpress site with Google Analytics by YOAST. I then added Google WebMaster tools through via verify through google analytics account. I then tried to set a preferred domain. I chose the non www. version; however, google wanted me to verify ownership of both versions in order to set a preferred domain. I then added the www. version of my domain. I was able to set the non-www. version to my preferred domain. Now, there are two example.com's in my webmaster tools. I have 10 sites. I intend to replicate this process on all of my sites. Do I have to leave the non-preferred version of my sites in the google webmaster? Can I delete it after I have set my preferred version? If I delete the non-preferred version will it delete my setting on the preferred version because it is now no longer verified (saved)?
Reporting & Analytics | | JML11791 -
Google Analytics Site Search to new sub-domain
Hi Mozzers, I'm setting up Google's Site Search on a website. However this isn't for search terms, this will be for people filling in a form and using the POST action to land on a results page. This is similar to what is outlined at http://support.google.com/analytics/bin/answer.py?hl=en&answer=1012264 ('<a class="zippy zippy-collapse">Setting Up Site Search for POST-Based Search Engines').</a> However my approach is different as my results appear on a sub-domain of the top level domain. Eg.. user is on www.domain.com/page.php user fills in form submits user gets taken to results.domain.com/results.php The issue is with the suggested code provided by Google as copied below.. Firstly, I don't use query strings on my results page so I would have to create an artificial page which shouldn't be a problem. But what I don't know is how the tracking will work across a sub-domain without the _gaq.push(['_setDomainName', '.domain.com']); code. Can this be added in? Can I also add Custom Variables? Does anyone have experience of using Site Search across a sub-domain perhaps to track quote form values? Many thanks!
Reporting & Analytics | | panini0 -
My own brand searches stopped appearing in google
Hi everyone, I've never asked a question here before, so go easy on me 🙂 I've noticed recently that my site is not showing up for my own brand searches anymore, and that my organic google results are way down. For example, it used to be if you would search for "grouvee kings quest 6" you would either see http://www.grouvee.com/news/kings-quest-6-retro-rerun-playthrough-part-1/63/ or http://www.grouvee.com/games/kings-quest-vi-heir-today-gone-tomorrow/69/ show up first in google. Now they're nowhere to be found. Here's another example. If you would search for "grouvee mass effect 3" you'd either see http://www.grouvee.com/games/mass-effect-3/75/ or http://www.grouvee.com/news/mass-effect-3-lets-take-the-earth-back/136/ show up first in google. Once again, nowhere to be found. I could go on with several other examples, but I don't really know what I've done. I 301 redirected 2 pages a few weeks ago because I added a /reviews/ structure to my site and needed to move a couple of pages that used to be under /news/ over to the new section, but those shouldn't have had anything to do with the pages I referenced above. I haven't done any link building really. I submit some of my articles occasionally to places like reddit or n4g, but that's about it. Anyone have any tips or things I should be looking at?
Reporting & Analytics | | petecorsaro0 -
Is the link data from Open Site Explorer in real time or an average?
I just started using Open Site Explorer to track internal and external link data. Is this information given in real time or is it an average over a specified period of time?
Reporting & Analytics | | mequoda0