· 6 minute read
The Impression Economy
A working proposal for valuing search visibility that does not lead to a click.
In SparkToro and Datos' 2024 study, roughly six in ten US and EU Google searches ended without a click. That does not tell us whether a particular brand appeared or was noticed. It does raise a question: what might click reports leave out?
A click tells you someone visited. It doesn't tell you everyone who noticed your brand. That distinction matters when you're deciding what your content is worth.
The great decoupling
On the Ahrefs blog, impressions and clicks began moving apart.
Impressions and clicks often get discussed together: a page appears, someone visits, and the report records both. It is tempting to treat more impressions as a promise of more traffic.
Ryan Law at Ahrefs found that the correlation between daily impressions and clicks on its blog shifted from +0.43 in the second half of 2024 to −0.35 in the first half of 2025. That is one site's experience, not a market-wide measurement.
Conceptual sketch of diverging trends, not a plot of the Ahrefs data.
Search Console records impressions for links and URLs, with rules that depend on the result type. An impression is not proof that someone noticed the brand. A brand mention without a link is not automatically a recorded site impression either.
If you only measure clicks, it's easy to dismiss impressions that don't lead to a visit.
That leaves out the possibility that someone noticed and remembered your brand.
You’ve always paid for impressions
Think about a yellow pages listing.
A yellow pages ad could help someone notice a business without calling immediately. That possibility was part of the reason to advertise. It did not mean every person who passed the listing remembered it.
Television offers a similar distinction. Seeing an ad and acting on it are separate events. A click is not the only possible sign of influence, but exposure alone does not prove influence either.
Display advertising gives us a familiar comparison: advertisers can buy opportunities for their message to be seen. That does not establish an equivalent price or effect for an organic-search appearance.
The question is whether some search appearances have an effect that a visit count cannot capture.
Evidence of influence without a click
A Google/Ipsos report published in 2014 described 61 simulated search experiments. Top-of-mind awareness averaged 8.2% in the control group and 14.8% in the group shown a top paid-search ad: a 6.6 percentage-point increase, or 80% relative lift.
The report found awareness benefits even without an ad click. Organic results were not changed in the experiment. It supports the possibility of influence beyond clicks, but it does not tell us what an organic impression is worth.
Repeated exposure can make a name more familiar. Whether someone remembers, trusts or chooses it depends on more than frequency.
Not all impressions are equal
A recorded URL impression, a named mention in an AI answer and human attention are different things. A model that compares them needs to make its assumptions explicit.
Proposed weighting model, not measured attention data. The relative weights need validation.
A named mention in an AI answer may be prominent. It can also be incidental or negative. I would not treat every mention as an endorsement.
A citation link gives someone a route to your site. Whether they notice it depends on how the answer is displayed and what they are trying to do.
A lower-positioned result may receive less attention, but its recorded impression still does not tell us whether someone looked at it.
A weighted impression model tries to account for these differences. It groups appearances by search feature and estimates how much attention each might receive. Those estimates need testing.
The relevance filter
The query matters as well as your position on the page.
What would someone associate your business with after seeing that result?
For a CRM company, appearances on CRM-related queries are more obviously relevant than appearances on a query about the time in Tokyo. The association still has to be noticed and understood.
Business relevance is another assumption to examine. Two companies with the same impression count may be reaching people with very different needs.
What your dashboard doesn’t show
A conventional report doesn't estimate the value of those appearances.
Hypothetical report, not results from a real business. Monetary values are estimates for illustration.
A report showing 5 million impressions and 150,000 clicks has a 3% CTR. That alone does not establish success or failure.
The arithmetic difference is 4.85 million. It is not a count of unique people who saw a brand without clicking, or a measurement of brand awareness.
One possible path is that someone encounters a brand, remembers it and searches for it later. Click reports may not connect those events. But rising branded search alongside falling organic clicks would not, by itself, prove that explanation.
Testing that hypothesis needs evidence beyond the two trend lines.
The silver lining inside the decoupling
The Ahrefs research also raises a question about the visits that remain.
In a 30-day sample reported in June 2025, Ahrefs found that visitors referred by AI assistants signed up at 23 times the rate of traditional organic-search visitors. Most of those AI signups came from ChatGPT.
That is one company's result, not an industry benchmark. It does not establish the conversion rate of AI Overview clicks or tell us what happens on another site.
The useful question is whether the mix and purpose of visits are changing on your own site. Fewer visits and better conversion can coexist, but neither should be assumed.
What this means for content strategy
If impressions have independent value, then the question for content teams changes.
It’s no longer just “how do we get more clicks?”
It’s “are we generating the right impressions?”
That means examining where your brand appears and what someone might learn from the appearance. Featured snippets, AI citations and other search features are worth observing, without assuming a universal attention ranking.
It means checking relevance, too. If most appearances concern topics far from your business, investigate whether they are helping you reach the right audience.
And it may mean bringing a scenario to a budget discussion: what comparison cost would different assumptions imply? Keep that separate from measured revenue, incremental return and brand recall.
This is a working framework. Its attention weights and relevance tiers are assumptions, not validated measures. A cost comparison can help examine those assumptions; it does not establish what an advertiser would actually pay or what your content has earned.
Your content may help people become familiar with your brand as well as bring them to your site. Click reports alone cannot establish the size or value of that effect.
The next challenge is to test whether these estimates explain anything that click reports miss.