Is your organic traffic eroding?
Over the last few years, increases in ads, SERP Features, and direct answers pulled away clicks from organic results, but Google doesn't make it easy to understand these changes.
Most importantly, Google recently announced making widespread changes to Search as more users interact with AI chatbots and platforms like TikTok to find what they're looking for.
Measuring the impact of SERP layout changes on organic traffic is complex. As a result, we often don't find the right way to turn negative traffic trends around and lose credibility.
The big question is, "How can you measure the impact of changes in the search results?"
The solution is an organic traffic erosion analysis.
What is SEO erosion?
SEO erosion occurs when organic traffic for keywords declines over time for a reason that's not related to an issue with the website.
An erosion analysis focuses on understanding three patterns:
Organic traffic moving to paid
Organic traffic declining due to non-paid reasons, like new/more SERP Features
Organic traffic decreasing due to fewer searches (market pullback)
The example in the screenshot below shows how organic rank is stable throughout the last 12 months, but CTR dropped.
The same view with impressions instead of rank (since we know it's flat) shows how impressions grow, but CTR remains unchanged. More people see the result but don't click because something else seems to deliver a better result.
3 reasons cause organic traffic erosion:
Top and Shopping ads
SERP Features like Featured Snippets, Instant Answers, Knowledge Panels, image carousels, video carousels or local packs
Query refinement bubbles take users down the funnel faster and on Google instead of other sites
The tricky thing about SEO erosion is that its impact is sneaky. You don’t notice it until your SEO revenue is down -20% Y/Y and you wonder what happened.
Erosion has been especially difficult to catch over the last 3 years because the pandemic disrupted user behavior so much. Many keywords have way lower search volumes this and last year because the pandemic pulled demand forward.
How to run an erosion analysis
3 red flags tell you if organic traffic erosion might occur:
If the organic position hasn't changed, but organic traffic declined, it's not an SEO issue but either ads or SERP Features.
If organic traffic/conversions declined and paid inclined, it's likely a shift from organic to paid.
If organic traffic declined and paid traffic was steady or declined as well, it's likely an issue of SERP Features.
We can use traffic data from Search Console and Google Analytics or your web analytic stool of choice to define and validate hypotheses for eroding organic traffic.
Just GSC data itself is already helpful to understand where you lost traffic without losing ranks.
But we can take it a step further. In the screenshot below, you see an example of an erosion analysis with changes in organic rank, traffic, conversions and revenue compared to the previous year and paid traffic. It allows you to quickly spot patterns of pages that lost organic but gained paid traffic or lost organic traffic with stable ranks.
Here's how you can build such an analysis yourself:
The first step to finding out whether your organic traffic is eroding or exploding is creating a custom click curve for the last full month and comparing it year over year (Y/Y). I've extensively described it in The importance of custom click curves, so I won't spend time on how to create one in this Memo.
An erosion has occurred if your click curve shows lower click rates compared to the previous year. However, click curves can't tell you why, just that something has changed. You still need to figure out the driver but be mindful that there might not be a single root cause.
To get closer to the main driver, you can use traffic data to narrow your hypothesis down. I'm using Google Analytics in combination with Search Console to describe the specific steps because it's the most used web analytics tool out there, but you can recreate the analysis with any tool.
In Universal Google Analytics:
Go to “Behavior → Site content → Landing pages”
Add segments for organic and paid traffic
Select the main conversion goals as event (requires that you tagged and define your main conversion and give it a value)
Next, get organic ranks for each page from Search Console. This step works best with the Search Analytics for Sheets integration because you get more data, but you need to run separate pulls for each year (the extension has no comparison mode). Make sure to select the same time period as in Google Analytics.
In the last step, use VLOOKUP or INDEX MATCH to match Ga and GSC data by URL and calculate the delta between the last month and the same month last year. You might have to remove the protocol (https://) and root domain (www.domain.com) from the GSC data to match GA data.
In the last step, sort the sheet by negative organic revenue or conversions and look at the columns for rank and traffic to define your hypothesis.
To validate your hypothesis, isolate pages that dropped in organic traffic and look at SERP screenshots for queries that lost the most clicks/CTR/impressions. You can get this in Semrush under Position Tracking or Keyword Overview and in Ahrefs under Keyword Explorer (compare SERPs between two dates).
The more you narrow your view down, the sharper it becomes. There are several ways to segment your data:
Compare one full month with the same month of the previous year
Narrow your view down to a subdirectory or even landing page
Look at data from only one country and/or device
Keep in mind GSC filters traffic out, especially the more URLs you have, so make sure to set up separate GSC properties for subdirectories
Filter out brand queries (only possible for GSC)
Prepare for more organic traffic erosion
AI Chatbots and the fragmentation of Search put a lot of pressure on Google. I expect more organic traffic erosion due to new SERP features, more people using AI chatbots to get answers and more users leaning on vertical search engines like Amazon, Pinterest or Booking to find better results.
To keep track of the changes, I don't only recommend running erosion analyses at least every 6 months but also storing rank and traffic data in a data lake and visualizing changes over time.
More about the impact of SERP Features on organic traffic: