TRY OUR PACKAGES
We learn from Mueller that the average position metric is based on real data. It’s a reliable measurement, even if it doesn’t always match up with what site owners see when they check their own ranking positions.
Google’s John Mueller answers a question regarding the accuracy of the average position metric which appears in the search performance report in Search Console.
The average position reported in Search Console may differ from what a site owner sees when looking up their own rankings in Google. Mueller says that could be due to factors such as personalization or geo-targeting.
Google’s John Mueller on Average Position Metric.
Read Mueller’s full response below.
Lastly, Mueller advises site owners to keep in mind that their URLs might be ranking in Google Images within the normal search results. Those ranking positions are taken into consideration when calculating average position.
Google Search Console tracks the average top position of a URL to calculate data for the search performance report.
“The data in the Search Console search performance report is based on actual search results that were shown to users. It’s not a theoretical number, but rather a number based on actual results.”
This question is answered in the latest installment of the Ask Googlebot video series on the Google Search Central YouTube channel.
“For web results keep in mind that your site might also be appearing in the images within the normal search results. In short, the average position and other metrics in the search performance report are based on actual search results, but actual search results can be quite varied.”
If multiple URLs from a website appear in search results, Search Console will use the highest ranking URL to calculate the average.
“The average position is based on the average top position of a URL from your site. If there are multiple URLs from your website which are shown in the search results page we’ll use the topmost one for this average. You might not always see the same position when you check yourself. That’s generally due to personalization, or geo-targeting, or because of short-lived visibility in search. You can sometimes make assumptions that this is happening if you see a number of impressions which is significantly lower than what you’d expect for those queries. That’s usually a sign that your site was only visible for a small part of the overall impressions.”
Google explains how accurate the average position metric is in Search Console’s search performance report.
Mueller begins his response by clarifying the average position metric is not theoretical. Like all metrics in the search performance report, average position is calculated using data from actual search results:
A Google Search Console help page further explains how average position is calculated using the topmost link to a page in search results:
To understand the average position number you need to look below the metric to see what is driving it. Google Search Console provides several dimensions we can use for this analysis — the pages, the queries, the countries, the devices, and the date. You can drill down and filter within the tool, but you can also export the data or use the API to perform different kinds of analysis.
Analytics Edge has a connector for Google Search Console that can download all of your data into Microsoft Excel. Combined with the Analytics Edge Core Add-in, you can automate your analysis and produce some insightful reports. For example, you can look at the distribution of the position for each page or query term [download this report]
The one thing you should not do is assume that the average position is a simple metric.
As an example, if one of your pages suddenly starts ranking (but not very well) for a brand new a query, you would think that would be a good thing. The problem is that the average position gets worse, not better, because the lower rank of the new queries pull it that way. It is even possible the new query term is not one you want to rank for.
Google Search Console makes it easy to track the position of your website pages in Google’s search results, but the numbers are averages, and averages don’t tell the whole story. When it comes to search engine ranking, they can be really misleading.
A Poor Key Performance Indicator.
It is tempting to think that, even if the number itself if not of much value, a trend over time may provide some insight — if it goes up or down, that would be a good indicator that things got better or worse. Many people use it in dashboards as a key performance indicator, and some regret that decision later. The problem with trending an aggregate position over time is that when something good happens, the number could go up OR down.
Or you could try to visualize the performance (click-through-rate and position) of various keywords for a specific page. [download this report]
The first problem is that the position metric is an average – a mathematical measure of central tendency. It is a good measure if the underlying data tends to be around a central value.
The problem is that a website is made up of multiple pages, and some page rank high and some low. For each page, some queries might rank high and others low. In some countries, a page may rank higher than in others for the same query. What ranks high on desktop may not rank well on mobile. Mix all of those together, and you can see why the average position for a website can be a meaningless metric.
Eventually you will end up with a different kind of summary report – one that suits your website. [download this report]