Aydın Nasuh You check your rankings with ShubHQ’s Rank Tracker. Your main page hasn’t moved much. Then you open Search Console and see fewer clicks.
It is tempting to rewrite the page, build more links or assume a competitor has passed you. But there is another possibility: the position is similar, while the search results around it have changed.
DataForSEO’s Search Insider 2026 report puts numbers on that problem. In its sample, desktop click-through rates fell sharply at the top of Google’s organic results. AI Overviews also became much more common.
Those are two observations, not proof that one caused the other. For a SaaS founder deciding what to publish next, both matter.
In the first post in this series, I looked at the websites Google AI Overviews and ChatGPT cite. This post is about the click you hope to get after someone searches.
The short version
- Rank is only part of the picture
- In the report's desktop sample, position #1 CTR went from 17.83% to 7.43% between September 2025 and August 2026.
- Comparisons face AI answers too
- AI Overviews appeared in 51.35% of commercial searches in the report's Q3 2026 panel, up from 26.30% in Q4 2025.
- Easy keywords aren't sheltered
- Keywords with difficulty scores of 1–20 still triggered AI Overviews 60.64% of the time in the relevant dataset.
- Measure the whole path
- Track impressions, rank, CTR, clicks and signups separately. A position alone can't tell you whether a page is helping the business.
A number-one ranking buys fewer clicks in this sample
DataForSEO’s organic CTR analysis used aggregated organic performance data from more than 2,000 websites to compare September 2025 with August 2026.
Desktop position #1 CTR fell from 17.83% to 7.43%. On mobile, it went from 25.17% to 20.80%. The desktop decline was much sharper.
Position #1 organic CTR, by device
Desktop
−10.40 ppMobile
−4.37 ppThat is a drop of about 10.4 percentage points on desktop. It is not a 10.4% relative decline. The distinction matters when you turn a research finding into a traffic forecast.
Here is an illustration, not a prediction. At 1,000 impressions, a 17.83% CTR would produce about 178 clicks. A 7.43% CTR would produce about 74. The ranking could look excellent in both cases.
But these are aggregate observations across different searches and websites. They do not show the same SaaS page holding the same position and losing that exact amount of traffic. Changes in query mix, search demand and the results page can all affect your own numbers.
The useful lesson is smaller: don’t use a fixed CTR assumption just because you reached a certain position.
AI Overview click share is a different measurement
Another section estimates where clicks go on results pages with and without AI Overviews. It uses clickstream-based estimates and excludes searches that produced no click.
Across four search intents, AI Overview citations received an estimated 42–45% of clicks when an Overview was present. The top organic result’s estimated share was 21–24%, compared with 33–38% on results pages without one.
Those numbers do not mean that 42–45% of all Google searches send a visitor through an AI Overview. Zero-click searches are outside this comparison. They also aren’t the same metric as the organic CTR figures above.
I would read this as evidence that the layout deserves attention. If Google places a useful answer and several source links above your result, your organic position tells only part of the story.
Being a cited source may create another route to discovery. It still doesn’t guarantee a visit, a signup or a customer. Measure each of those separately.
Product comparisons aren’t outside the change
It is easy to think of AI Overviews as a problem for simple definitions: a question gets answered, and the reader leaves.
The report’s intent breakdown reaches further. Informational searches had the highest trigger rate, but commercial searches also saw substantial growth.
Searches that triggered an AI Overview
Commercial trigger rates rose from 26.30% in Q4 2025 to 51.35% in that partial Q3 2026 period.
For a SaaS founder, comparison and alternative queries deserve a manual look. The aggregate commercial rate isn’t a measured rate for SaaS comparisons. It is a reason to inspect your own results before assuming those searches will send the traffic an old forecast promised.
Ask what the searcher still needs after reading the Overview. A real product walkthrough, a specific limitation, original test results or a useful comparison may give them a reason to continue. Another generic introduction may not.
That is my editorial judgment, not a content formula tested by this study.
Low difficulty doesn’t mean an open lane
When you have a small site, low-difficulty keywords are an understandable starting point. The problem comes when “easier to rank” quietly becomes “likely to bring clicks.”
In the report’s keyword-difficulty analysis, the easiest bucket, KD 1–20, triggered AI Overviews 60.64% of the time. The hardest bucket, KD 81–100, had a lower rate of 53.76%.
That doesn’t make difficult keywords a better choice. It shows that ranking difficulty and AI Overview presence describe different things.
Low volume wasn’t a reliable escape either. In the quarterly panel, keywords with 201–300 monthly searches had a 58.59% trigger rate in Q3 2026. The 601+ bucket was lower, at 48.62%. That panel excludes keywords below 100 monthly searches, so it cannot settle what happens across the smallest long-tail queries.
I would still look for specific buyer questions. I would just check the actual results page before putting a week into the article.
What I’d check before rewriting a page
A decline in clicks is a symptom. It doesn’t tell you which part changed.
A practical check in Search Console
Start with one page and its queries
Compare equivalent periods. Look for changes in which searches generated impressions instead of relying only on the page total.
Separate desktop and mobile
Check impressions, CTR and average position for each device. The report found different patterns, and your own combined average may hide one too.
Read the live results
Search your important buyer questions in the relevant country and language. Note AI Overviews, ads, videos and the sites above your page. A manual check is a snapshot, not every user's experience.
Check the business result
Compare landing-page visits with signups and paying customers where your analytics supports it. Fewer visits and fewer customers aren't interchangeable findings.
Choose the fix that matches the evidence
A ranking loss, lower demand and lower CTR with similar visibility need different responses. Don't rewrite a page solely because its click total fell.
Average position also needs care. A similar page-level average can conceal different queries, countries or devices moving in opposite directions. “The ranking held” is a hypothesis to check at query level.
How I’d choose the next SaaS article
My next article would start with a question close to a real product decision. Then I would inspect the result, identify what remains unanswered and decide whether I can add something useful from experience.
For ShubHQ, that means connecting the buyer question, the competing tools and the evidence I can actually show. I wrote about that connection in the SaaS positioning post.
I would keep a short record for each target query:
Before you commit to the next article
- The question belongs to someone who could use your product.
- You checked the results in your target country and language.
- You recorded whether an AI Overview appeared.
- You can explain what your page adds beyond the answer already shown.
- You have a way to measure visits and signups after publication.
- You will judge progress using your own baseline, not a universal CTR benchmark.
Search Console can tell you how your own Google visibility and clicks change. Use ShubHQ’s Keyword Tracker to track your target keywords’ Google rankings alongside those numbers. A separate AI visibility check can show whether your product is named and which sites are cited. Neither replaces the other.
You can run ShubHQ’s free LLM visibility checker for an initial check, or create a free account to track your category’s buyer questions over time.
FAQ
Can organic clicks fall without a large ranking drop?
Yes. Clicks also depend on impressions and click-through rate. Check query mix, device and country before treating a similar average position as proof that every ranking held.
Does this research prove AI Overviews caused the CTR decline?
No. DataForSEO explicitly says the organic CTR analysis does not establish whether AI answers, results-page changes or user behavior caused the decline.
Do AI Overviews receive 42–45% of all Google searches?
No. That figure is an estimated share of clicks on results pages where AI Overviews appeared. The comparison excludes zero-click searches.
Are low-difficulty keywords protected from AI Overviews?
No. The report found a 60.64% AI Overview trigger rate for keywords in the KD 1–20 bucket. Difficulty and AI Overview presence measure different aspects of an opportunity.
Sources and limits
- DataForSEO: Search Insider 2026 Edition, October 2026. Organic CTR: printed pp. 9–11; estimated click distribution: pp. 11–12; search intent, volume and difficulty: pp. 13–15; methodology: pp. 31–32.
- The organic performance sample covers more than 2,000 websites, September 2025–August 2026. The click-distribution analysis uses a separate keyword dataset and clickstream-based estimates. The quarterly SERP panel has its own coverage and excludes keywords below 100 monthly searches.
- The report’s primary search analyses focus on US, English-language results. These are broad findings, not a SaaS-only sample or a forecast for your site. The workflow and editorial suggestions above are my interpretation.
- More from this report: 7 numbers for SaaS founders.
