Is Google Search Console Rank Tracking Data Accurate?

Google Search Console shows an average position for all queries run by all the searchers within the selected timeframe. You can see the average position for all keyword phrases combined or just for the selected ones. In that context, this data is accurate.

Is Google Search Console good for rank tracking?

The short answer is…

No!

Surprised? Let me explain.

First, we need to clarify what it means that rank tracking data is accurate. Google is informing you on which positions your website has been displayed to the users when they were typing their queries into the search box. The only situation when this data could be considered inaccurate is if Google would display different results when a user performs an actual search and different ones when you look into the Search Console. They are basically showing you the data they already own. Although there were certain concerns about the reliability of this data I think we can safely assume it is more or less accurate right now.

So what is the problem, you say?

The key is this part of the official Google explanation:

A link must get an impression for its position to be recorded. If a result does not get an impression—for example, if the result is on page 3 of search results, but the user only views page 1—then its position is not recorded for that query.”

So, let’s say we had 10 searches for a certain keyword today. Let’s call that keyword The X Keyword (mysterious, I know…). For 9 of these searches, your website was ranked somewhere on the 3rd page of results. But no one opened even the second page! So positions from these 9 times were not recorded. For one last search, we had a guy – call him Chris. Chris is familiar with your website. He landed on it a couple of times before and spent some time reading your content. Google knows that. In addition, before typing our phrase, Chris typed a few other, similar queries and was not satisfied with what Google returned. He rephrased the question and ended up entering our X Keyword. Google has filtered out a few results from his previous queries, moved up your page a bit because they know he likes it…

…and Boom! You are now on the second place! For that one user, for that one search.

What will Google Search Console report as your average position for this day? A second place of course. Is that what you expect from a rank tracker? I guess not. 

Note!
What you probably expect is a depersonalized ranking. This is a more objective way to assess where Google is putting your website most of the time. Or at least calculate the starting point (before Google takes personalization factors into account) for each keyword. Of course, there are certain personalization factors that you may want to include when you track rankings. These are typically country, language, sometimes exact city, or a user device. You can set them up in most modern rank trackers.

If you are looking for a free tool to track your rankings, check out my list of free rank trackers.

I had many cases like this, maybe not all of them were that spectacular but still – the differences were significant. The worst part is you don’t have a way to estimate an error because for each keyword it is different. For keywords outside the Top10 your data may be skewed by default. For keywords within Top10, it is most likely closer to the truth. But again, even if Google Seach Console shows that your website is in Top10 you have to cross-check this with search volume for the keyword and your number of impressions. Only this way you can verify if your website is close to the presented position or is it like in the example above.

How automated queries mess up with Google ranking data

You thought that is all? Nope.

I have something even more interesting for you. The real fun begins when we add automated queries to the mix.

Wait… what?

Because Google does not provide a reliable way for rank tracking, there are dozens of third-party tools for that on the market. Besides rank trackers, we have keyword research tools and various search engine scrapers. All of them combined are executing millions of Google queries each day.

Yes. Millions. Each day.

How this affects the data in Search Console? Take a look at below screenshot:

This is a view with GSC ranking data for a single keyword. You can see that within a four-day span I had a total of 78 impressions and my average position was 73.3. Remember that “a link must get an impression for its position to be recorded“, right? So it means that for this particular keyword there were 78 users who went to at least 8th page of results. All that in the four-day span? Even if you assume that this average position is a result of some weird mix where for half of the searches my page was in Top10 and for the other half, it was somewhere on the 15th page it still looks a bit odd, doesn’t it?

Yeah, but what if this is a keyword with tens of thousands of searches per day? Maybe every once in a while, someone goes through more pages?

No, it is not that kind of keyword. To be honest, I do not believe it has more than just a couple of dozens of searches per month.

What is the explanation then?

Automated queries. Most of the tools responsible for those queries scrape much more than just the first 10 results. The keyword that I used in this example is likely to be monitored this way. Google does not like automated queries and they are pretty good at filtering them out. Or at least in reducing their number. But for the reasons that God Google only knows, this data sometimes lands in Search Console rankings. This makes it even more skewed when it comes to rank tracking.

So, Google Search Console is a very useful tool with lots of valuable data. But if you are looking for something to track your rankings it is not the best choice. In such a case read my article about free rank trackers.