Table of Contents
- How Impressions Are Measured in Google Search Console
- Identifying When SERP Features Influence Impressions
- SERP Features With High Influence on Impressions
- Key Takeaways
Have you ever reviewed data within Google Search Console to find a sudden and significant spike in impressions, where clicks don’t follow, without an obvious reason for the change?
This situation is very common within GSC properties for sites with any kind of meaningful data. In fact, this is by design in GSC, where the data you’re seeing is most likely intended to be presented in that way.
Within my post, I’ll be detailing five situations for high impressions and low or no clicks at all that I often come across when working with large-scale sites across various industries.
In particular, my post will first provide an outline of how impressions are measured in GSC, tips for knowing when SERP features have influenced impressions, five different reasons for a spike in impressions with few clicks, along with some key takeaways.
How Impressions Are Measured in Google Search Console
When reviewing impression increases in GSC (where clicks don’t follow), it’s important to highlight that we’re purely looking at the Performance report in GSC for ‘Search results’, not Discover or Google News.
If you’re seeing an increase in impressions, we need to then understand the definition of an impression based on Google’s documentation and what that means for the data that you’re looking at. Google’s definition of an impression is:
How often someone saw a link to your site on Google. Depending on the result type, the link might need to be scrolled or expanded into view.
Now, this is a very broad definition of how an impression is calculated, where many jump to the conclusion that the data that you’re looking at the Performance report in GSC is just for a ‘plain blue link’ (how Google defines standard organic results).
The reality is that plain blue links aren’t the only feature in web search (the main section of Google) for how impressions are triggered in GSC. There are various other features, which are designed to garner a lot of impressions, but are far less likely to be clicked on.
Identifying When SERP Features Influence Impressions
Within the GSC Performance report, it’s not possible to filter by every single SERP feature. The only data we have access to are the result types or features in the Search Appearance filter.
Within this filter, we currently have access to data such as FAQ rich results, Product Results, Review Snippets, Web Stories, Job Listings, and more recently Translated Results.
Four out of the five situations covered in my post don’t have the ability to surface within the Search Appearance filter in GSC, which is the primary reason for the confusion that is often experienced by SEO professionals.
To understand which exact SERP feature has delivered the increase in impressions in GSC, there are some clues we can use to get to the bottom of the sudden spike. The clues are the data in GSC that accompany impressions: the query, average position, device, and country in particular.
Using each of the four data points mentioned above, we can try to reverse engineer where the increase in impressions is coming from by using 3rd party tools to look back in time.
In the many cases, it’s unlikely that you’ll be tracking the queries in 3rd party SEO tools, so there won’t be historical ranking data to reflect on to see what the SERP looks like. There are various tools that can give you historical SERP data, but only one is truly helpful for this situation.
The tool that I highly recommend for troubleshooting sudden impression spikes is the Semrush SERP screenshots tool. Up to 90 days of screenshots can be accessed for high volume queries (relevant to impressions spikes) that aren’t tracked in a campaign, which is incredibly useful.
SERP Features With High Influence on Impressions
Now that you have gained a better understanding of how impressions are gathered within Google Search Console and the process for identifying when SERP features are of influence, we can get stuck into the reasons for high impressions and low clicks.
Here are the top five reasons, based on likelihood of triggering an impression in GSC without barely any or zero clicks being recorded within the property.
1. Internal Sitelinks
The first situation is the most common out of the five. Sitelinks can be generated algorithmically when related content is available for a plain blue link. There are various types of organic sitelinks, with the sitelink type in question being the ‘internal’ variation.
Within the internal sitelink variation, there are two subtypes which can either show alongside branded searches and a variation which looks like any other sitelink, showing in the simplified format underneath the snippet.
When internal sitelinks are triggered for a web page, related content will then start to garner impressions at the same rate as the ranking result. So for every time the internal sitelinks appear alongside the result (based heavily on the query), an additional impression is recorded.
Here’s an example of what this looks like for a page that is yielding internal sitelinks for the search term “seo” when searching from the US, employing the standard internal sitelink treatment (as opposed to the branded variation):
For every instance that the plain blue link appears (the result with the snippet), the internal sitelink URLs will also trigger an impression when scrolled into view. This means the ‘Introduction to Indexing’ page will also have impressions for “seo” when visible.
As mentioned, this example shows the standard internal sitelink treatment. For the branded internal sitelink treatment, this can have a noticeable impact on impressions if the brand name gets a lot of searches on Google. Here’s an example of how this can show in GSC:
In the example shown above, filtering has been used by query and landing page. When selecting the ‘about’ page of the site for the brand name, we can see sudden spikes in impressions recorded in GSC.
Spikes like this can happen on both mobile and desktop devices, and can be short-lived depending on how useful the URL is to users. Because the ‘about’ page is added and removed suddenly from the internal sitelink treatment, there must be other pages that are more relevant.
As mentioned, internal sitelinks are the most common contributor to sudden spikes in impressions that have low clicks. Though they don’t create as much confusion as the features to come, or generally result in the same discrepancy among impressions and clicks as others.
h/t to Lee Foot on Twitter for this addition to my list.
2. Images in Image Packs
Images have the ability to surface within a SERP feature on Google called an Image Pack. The Image Pack appears in web search, and has the ability to influence the Performance report in the ‘Web’ Search Type, not just the ‘Image’ Search Type.
It is the ‘Web’ Search Type filtering that we’re interested in for this post. With the Image Pack feature being used on 27% of mobile SERPs and 16% of desktop SERPs according to AWR.
The example below shows a very common situation that results in a significant amount of impressions with very few clicks. The second image that appears in the Image Pack is from my own site, and is consistently being triggered in the SERPs for various countries (confirmed through GSC data).
Again, this is data that will appear with the GSC Performance report under the default ‘Web’ Search Type filter. This will be surprising to many, as you may think that this data should be located under the ‘Image’ Search Type filter, which is only for data from the Image Search tab on Google.
Because this is data taken from my own site, I’m happy to show how this is represented within GSC to assist with understanding. For the search term “featured snippets”, here’s how this data is represented within GSC for my site:
The impressions are quite consistent over time for my image, which has only garnered 9 clicks from almost 70K impressions. This is completely normal and how this data should be represented in GSC, so don’t get too attached to your average CTR calculation for your site in GSC if you rank within a lot of Image Packs on Google.
Image Packs appear on both mobile and desktop search results, so the data should be reflected in this way when appearing in Image Packs. I can see that most of the impressions for my image were on desktop, with all clicks (a grand total of 9) all coming from desktop also, which is telling to how users search on Google for information about Google’s Featured Snippets.
3. Images in Knowledge Panels
Images appearing within Knowledge Panels are another instance where images in web search can trigger a spike in impressions. This is another example where the average position metric by device type for a query can hint at the feature, something that’s covered in Google’s documentation.
According to Semrush, Knowledge Panels appear on 26% of desktop SERPs and 28% of mobile SERPs, being a feature that appears a similar amount of time as Image Packs.
In the example shown below, an image from my site is surfacing as the first result in the Knowledge Panel for the query “snippet”. And based on the data, which I’ll show in just a moment, it looks like images from several different pages have shown in this KP in the past.
Similar to the Image Pack situation, the impressions from the images (well, the URL that the image is located on) are all collected within the Performance report in GSC. For the first image that is shown above, this is the one with the most impressions and has an average position of 10.9 on mobile and 14.2 on desktop in the US.
Here’s what this data looks like in GSC which is actually spread across 13 different URLs, but highly concentrated around the first two images shown above which has historically shown within the KP more consistently compared to images on other pages:
Interestingly, in this example the images are being alternated over time, but the impressions have remained consistent. This gives me confidence that at least one image from my site has shown within Google’s KP for the query “snippet” consistently over time.
Knowledge Panels are a feature that appear on both mobile and desktop search results, with the CTR being exceptionally low, which can be quite confusing for site owners if they don’t understand where the visibility is coming from. In the example I’ve shown above, my images have received 671K impressions with only 17 clicks over the past 16 months.
Tip: Glenn Gabe has written some great articles on how images in both Image Packs and Knowledge Panels are represented in GSC. Along with a more in-depth look at how measurement in general works in GSC.
4. Links in Tweet Carousels
Links within tweet carousels are likely to be the most confusing for site owners to understand out of the five different reasons for high impressions and very low or zero clicks in GSC.
This is why I conducted a case study last year on the influence of twitter carousels on GSC data. The experiment involved my own Twitter carousel by sharing a link and getting hundreds of SEO professionals to view and click the link within my tweet.
The result was a ton of interesting data collected within GSC for my site, and also in Google Analytics. So if Search Engine Land were to share a link in a tweet, and a user were to search on Google and click the link directly from Google’s SERPs, this would all be recorded in GSC.
According to Semrush, Twitter carousels appear on 8% of desktop SERPs and 7% of mobile SERPs. But my experience is that this is quite a rare situation for site owners to come across, considering it’s mostly for broad queries (not peoples names) are the queries where the impression spike comes from, but it’s seems to be more rare these days for Google to show Tweet Carousels for broad queries.
Aside from the screenshots shown in my experiment, I don’t have an easily accessible example of Twitter Carousels causing a spike in impressions in GSC, but I’ve certainly seen examples in the past where the impression spike leaves site owners confused.
Twitter carousels are a feature that appear on both mobile and desktop search results, and when a link is shared within a tweet, that link can yield impressions in GSC if the item within the carousel is within view on page one. It is however exceptionally rare that a Google user would then click the link itself from the tweet, then resulting in a short-lived impression spike in GSC.
5. Web Stories in the Visual Stories Unit
Google’s Visual Stories unit is one that has come about more recently for Web Stories. Google increased prominence of the Visual Stories SERP feature considerably in April of this year, resulting in sites who were creating high quality Web Stories to receive a large increase in impressions.
Much similar to the four other examples covered in this post, Web Stories that appear in the Visual Stories unit receive a ton of impressions, with very few clicks. This may be because Google is still testing where they fit best in search, so are showing them for overly broad queries such as “seo” and “search engine optimization”, which my Web Story ranks for both in the US.
Here’s what this can look like for a site when my Web Story is included in the Visual Stories unit, that appears exclusively on mobile search results:
You would think that ranking for the search term “seo” (on the first page of Google), would result in a good amount of traffic to whatever URL that is showing, but this does not appear to be the case, and I’ve had this experience corroborated by other creators such as Andrea Volpini.
As mentioned, ranking for the query “seo”, “search engine optimization” and even “seo tools” on mobile in the US on the first page is great. But the feeling is diminished when looks at the data in GSC, which shows the following for the search term “seo”, which I’ve ranked for consistently since the Visual Stories unit started showing more often on mobile in the US:
As you can see, there was a significant overnight spike that happened for my Web Story URL once it was added to the Visual Stories unit on mobile in the US. It is however important to note that the spike in clicks that followed was because SEOs were searching for my Web Story with the query “seo” after I posted about it, so you can ignore that spike.
According to AWR, the Visual Stories unit shows on 3% of all Google search results on mobile in the US at the time of writing this post. Google isn’t giving up on Web Stories just yet like they did with AMP, so I would expect this figure to increase over time and for the feature to expand to other regions (not just showing in the US).
The helpful aspect to Web Stories is that there is a specific Search Appearance filter that you can use to assist with reproducing an impression spike, unlike the others mentioned in this post. The ‘Web Story’ filter within the Search Appearance tab shown above will show basically all of the impression data in the graph, providing another clue to the fact that all of the traffic was mobile-based and because of the Visual Stories unit.
Significant Impression spikes without clicks in Google Search Console can be a confusing situation when you’re not aware of the impact that some SERP features can have. My post covered the top five features where I’ve witnessed confusion from site owners, ranging from features such as Knowledge Panels (where images can appear) to Twitter Carousels.
Within my post, I covered various takeaways that are important for site owners and SEO professionals to keep in mind when dealing with impression spikes in GSC that don’t involve any (or very little) change in clicks. Here’s some of the top takeaways from my post:
- The definition of an impression in GSC tends to focus on URLs that are either scrolled into view or expanded in a way that makes the URL visible to users.
- It’s important to be aware of the Search Appearance filters that can appear in GSC. If there isn’t a filter triggered that correlates with a query and landing page, then it could relate to one of the four SERP features discussed.
- When troubleshooting the feature that has triggered the impression spike, if you’re unable to use a VPN (because the spike was short-lived and happened in the past), then you should be using the Semrush SERP screenshots tool.
- The five different situations detailed in my post include Internal Sitelinks (there’s branded and non-branded types), Images in Image Packs, Images in Knowledge Panels, Links in Tweet Carousels, and Web Stories in the Visual Stories unit.
- Each of the five features can have varying levels of impact on impressions in GSC, with some having a short-lived spike (such as the Twitter Carousel feature) and others being more long-lasting (such as the Images in an Image Pack).
By being aware of the five different reasons for an impression spike (with barely any, or zero clicks) you’ll be one step ahead of others when reviewing your GSC data. Knowing the source of the data helps give confidence to your clients that you know the source of visibility of their site, instead of pinning it down to a bug or an issue on Google’s end.