Editor’s note: This post was updated on July 8, 2022, to add new data about keyword information missing from Google Search Console (GSC), how GSC measures impressions and areas where Google Search Console (GSC) and Google Analytics differ in their reporting.
Google Search Console data can be misleading
I’d like to walk through how the data Google gives you is misleading and why I recommend that my clients use ranking data from an SEO software platform like Ahrefs, SEMrush, or GetStat instead for their SEO measurement and planning.
One caveat before we get started…
I think it’s incredible that Google gives us (as web admins/marketers) free data so we can see how our website is performing in Google organic search. Not only are the technical reports helpful for diagnosing traffic decreases and SEO issues (like issues with your AMP pages), but the information Google provides around performance is the ONLY place you can get that information.
I also like that Google says that they are always looking for feedback to improve Google Search Console (GSC), and if you’d like to provide feedback, you can do that here (inside the interface):
As always, your feedback is important to us! Let us know in the main menu what's working for you, and what we can improve in the new Search Console. pic.twitter.com/OOcWq9tMHa
— Google Search Central (@googlesearchc) December 13, 2018
With that said… Let’s get started, shall we?
I’m going to cover the following issues with Google Search Console data:
- Google hiding keyword terms for 46.08% of clicks
- Misleading impressions and ranking data
- Which data is not tracked
- Consolidation of URLs – only reporting on canonical URLs
- Limitations to the API data from the new Google Search Console
- GSC webmaster notifications are misleading
- GSC will never match GA data
- Google Search Console doesn’t filter out bot traffic
1. Missing keyword term data
Patrick Stox, a Brand Ambassador for Ahrefs (June 2022), ran a study over one month looking at data across 146,741 websites with almost 9 million clicks and found that Google often hides keyword data from site owners. – they even used the API, which should have allowed them to get all the data from these sites.
In his analysis, Stox notes, “You see lots of sites around the middle and a large spike at 95%-100% missing clicks. So many of the sites are missing about half their data, but a large number of sites are missing most of the data.”
Google states that they hide keywords for privacy reasons and when the queries are only made a small number of times. However, these sites (and maybe many of us) are missing long-tail queries that Google is now showing us. In addition, they created a Google Data Studio report you can use to see how much of your data is missing.
2. Misleading impressions and ranking data
For each dataset in the Search Analytics report (which reports on the keyword, URL, country, device, and search appearance of your URLs), Google will provide metrics on the average ranking, number of impressions, clicks, and click-through rate. BUT, Google Search Console only reports data when it’s hit a minimum number of impressions, and Google does not specify the exact number where you’d meet the minimum. That means GSC is unreliable when looking at low search volume terms based on Moz experiments.
As a result, both impressions and clicks in GSC reporting are not reliable enough to be used for SEO planning.
Here’s how GSC measures an impression:
The URL is considered a viewed result, even if your URL is not scrolled into view
Here’s Google’s confusing explanation.
GSC measures impressions by the “number of times any URL from your site appeared in search results viewed by a user…whether or not the item is scrolled into view, as long as the user need not click to see more results (such being required to click “see more” to see the link).”
Except for expanding or scrolling search widget where “the item must typically be scrolled into view within the carousel, or expanded by a click, to register an impression.”
Impressions in the Performance Report do not include PPC impressions, Discover, or Google News.
However, it does include your URL appearing in the following SERP features:
This can add to the confusion in your data, as this is why “current issue” is “ranking” for the term PNAS – because the link appeared and received impressions when the user queried “pnas”. Often this results in those URLs having high impressions but low clicks.
Images in Image Packs
You would think this would be recorded in the “Image” Search type in Google Search Console, and it IS recorded there, but it can also appear in web search, so that means it influences the Performance report.
Both of the red boxes below are URLs on pnas.org:
Images in Knowledge Panels
Google covers this in their documentation, but if you have a URL appearing in a Knowledge panel, you’ll see those impressions in your Performance report.
See the example:
Links in Tweet Carousels
If your URL is shared in a tweet carousel and is visible on the page (even if the searcher needs to scroll for it to be in view), it will receive an impression and be recorded in the performance report. Here’s an example:
These generally only appear for broad topics on mobile and only in a small percentage of searches.
Check out Brodie Clarke’s post for a full deep dive into how impressions can be misleading.
The data is aggregated and averaged
The count is aggregated by property or page. As a result, you’re seeing the total number of impressions across the time period.
Sometimes your asset needs to be scrolled into view to be measured
Unlike impressions on search results pages, infinitely scrolling result pages (like image search) might require the item to be scrolled into view for the impression to count. Learn more.
Again, Google’s illustration:
I particularly like the disclaimer on the top of Google’s instruction page that mentions that the requirements, like items needing to be visible in the carousel, etc., are subject to change. So it’s hard to feel confident in the metrics you’re looking at, especially when the rules are likely to change at the drop of a hat.
And the ranking position is only recorded once an impression for the URL is recorded.
Here’s how GSC measures rankings or average position:
It’s an average
Google defines the average position as the average of the top result from your site for that keyword. So, if your site has 3 URLs ranking at positions 2, 4, and 6, the position is reported as two over that time period.
Let’s talk about the impact of GSC reporting an average position for your URL over the time period you select (say one month). What if that page suddenly started ranking for a term that wasn’t very relevant on day #4? Or did it rank poorly for a term during that period? That would skew your entire average position for that month.
If your URL(s) are in a Knowledge Panel with multiple links, they will all be recorded as the same position.
Just like Google Analytics, if you select the time period without understanding how the day of the week or seasonality can hugely impact your metrics. Considering these variables can skew what data you think you’re reviewing.
Rankings change hourly
Since average position ranking is based on the time period selected, another factor to consider when looking at the data is that rankings can change by the hour. Here’s a great article about the impact hourly rank tracking can have on your “average” accuracy.
Imagine this scenario: You add new content. Your keyword starts at position 80, moves to 70, 60, and eventually to #1.
Imagine you create a different piece of content, and it sits at position 40, never wavering.
GSC will report both as having an average position of 40. However, at this point, that metric is unusable from a strategic perspective.
Click data discrepancies
Often the click data you have in Google Search Console will not reflect the click data you have in Google Analytics for that same page and time period. There are a variety of reasons why this is, but I’ve highlighted the more significant reasons below:
Let me explain why 301 redirects can cause data confusion. (I pulled the below quote from the same Moz study above):
GSC tracks a click based on the URL in the search results (let’s say you click on /pageA.html). However, let’s assume that /pageA.html redirects to /pagea.html because you were smart and decided to fix the casing issue…
If Googlebot hasn’t picked up that fix, then Google Search will still have the old URL, but the click will be recorded in Google Analytics on the corrected URL, since that is the page where GA’s code fires.
Personal Identifying Information (PII) data
Sometimes Google Search Console doesn’t show the keyword data because it’s too specific and would be personally identifiable.
Google Search Console data has a lag
Google Search Console says that collected data is available in 2-3 days, which is a significant lag. So if you’re looking at Google Analytics data from yesterday, you’ll never be able to match it up with GSC data.
Time zone differences
While you can set Google Analytics to record data based on EST, GSC is set to PST time, and their day would therefore be labeled according to the local time in CA.
3. Data elements that are not tracked
It turns out that there are Google elements that appear in search and link to your content that you cannot track in Google Search Console or Google Analytics. I’m just going to highlight here the top-level elements that are currently not trackable in Google Search Console (for a deep dive, I’d encourage you to read Glen Gabe’s post):
In the Google Search App:
More like this:
Video results elements
New video packs:
Google SERP Features:
Featured Snippets with faceted navigation:
People also ask:
4. Consolidation of URLs – only reporting on canonical URLs
A heads up as you move forward with your monthly SEO reporting. If you suddenly see URLs that don’t make sense, it’s due to Google’s announcement of how it’s only tracking the Google-selected canonical URL. This URL may or may not be the URL that you want Google to use as the canonical. Google is also ignoring URLs that might have added UTM tracking. This tracking method makes it difficult to determine performance by page.
Based on the SEO audits I work on, the canonical Google selects may or may not be the URL that would consider YOUR canonical URL.
5. Limitations to the API data from the new Google Search Console
Say you have a bigger site. Even with the restrictions we’ve mentioned above, you want to pull your data into a Google Data Studio dashboard, but the 1,000 rows of export are restrictive (and manual).
With the new GSC, you can now get 16 months (486 days) of data. This newer feature is an improvement as previously; it was only possible to pull around three months of data- what was visible in the interface.
However, with the API, you can get 5,000 lines of data per query.
Why is this important? If you run an enterprise-level site, you still can’t get ALL of your data. You’ll still be missing landing pages that are potentially ranking for important search queries – all items you can track with an external SEO tool.
If you have additional data that you’d like to see available via API, Google is requesting feedback here:
👋 site owners, SEOs, and developers! What Search stats or functionality is your 🥇 most wanted to get via API? We'd love to hear what you'd do with it — let us know in this short form ⬇️⬇️⬇️ https://t.co/PmiRt00qLb
— mariya (@marrrr) March 18, 2019
6. GSC webmaster notifications are misleading
Recently, Google started sending out notifications to website owners whenever there is a drop in impressions and/or site clicks. This might seem like a great idea, but…
Google adjusts the algorithm and ranking of various keywords ALL the time. Often a site loses a particular ranking because Google has gotten smarter about what content exists on the site. Google might then offset that drop in impressions with an increase on another keyword or landing page.
But, if a website owner receives this notification every time there is a drop in impressions and clicks without any other information, it could cause undue stress:
Regardless of intention, this notification is sure to cause undue webmaster stress.
7. Google Search Console (GSC) data will never match Google Analytics (GA) data
Because the two Google services were built to address two different purposes, they aggregate their data differently.
Here’s a great infographic showing the difference between the two:
In Google Analytics, the Search channel also includes Google Discover. In Google Search Console, that is a separate report.
Some users block Google Analytics data collection, but there is no block for Google Search Console data.
Google Search Console is delayed in its reporting by several days, but Google Analytics has data within seconds of a page being viewed. It’s also set to Pacific Standard Time, where you can set your Google Analytics to your home time zone.
Landing pages in Google Search Console are aggregated based on the canonical that Google chose for you to show in search. And in Google Analytics, you can see every landing page visited from search whether or not it’s the Google canonical.
8. Google Search Console does not filter out bot traffic
Yes, really. That spike could be a bot searching for odd keywords and clicking through the SERPs to your client’s side. (Source)
As you can see, there are a few areas where data from GSC might be misleading and will not be reporting what you would expect.
If you’re running an in house SEO program, I often recommend that my clients purchase an SEO tool to help them measure rankings and I recommend different tools based on their business needs. If you need to get started evaluating SEO tools, check out my blog post of the top 3 in the market (Moz, SEMRush, and Ahrefs).
Have you seen GSC data discrepancies? I’d love to hear about your experience in the comments!
PS: Thanks for taking the time to read my post and geek out about SEO with me! I get my inspiration for post topics from other SEOs and in-house marketers struggling with SEO strategy and implementation questions, so if you like this post, please…
1.) Leave a comment
2.) Share it with your network
3.) Follow me for future posts