Search traffic used to be predictable: rank well, get clicks, convert some of them into trials. That model is breaking for software companies specifically, because the buyers researching a new tool are exactly the audience Google now tries hardest to answer without sending anywhere.
For SaaS and startup teams, the future of AI search already shows up in their own analytics: flat rankings paired with falling clicks.
What AI search is doing to SaaS traffic right now
Pew Research Center spent a month in 2025 tracking what actual US adults did after they searched Google, using consented browser data instead of a survey.
When an AI summary showed up, Pew found that people clicked a traditional search result in just 8% of visits, against 15% when it didn't.
The gap doesn't stop at the click. People who saw a summary also ended their whole browsing session more often, 26% of the time versus 16% when they didn't.
| Search results page | Clicked a traditional link | Ended the session after one search |
|---|---|---|
| With an AI summary | 8% | 26% |
| Without an AI summary | 15% | 16% |
The future of AI search points toward less traffic, not more
This pattern predates any single study. Search Engine Land reported that Gartner, in a February 2024 forecast, expects traditional search engine volume to fall 25% by 2026 as more queries get answered by chatbots and other virtual agents instead of a results page.
Gartner's Emily Weiss told marketing leaders whose brands depend on SEO to start testing other channels, part of a broader push toward diversified growth marketing rather than waiting the shift out.
By June 2026, Search Engine Land reported, an analysis of Similarweb clickstream data by SparkToro found that US Google searches ended without any click 68% of the time, up from roughly 60% two years earlier.
Zero-click search is turning into the default outcome, not a footnote. We looked at what that shift means for pipeline specifically in our earlier look at SaaS startup AI Overviews.
How SaaS and startup teams are actually responding
Photo by StartupStockPhotos on Pixabay
The response is already funded. Berlin-based Peec AI raised a $21 million Series A in November 2025 to build a dashboard that tracks whether a brand shows up when someone asks ChatGPT and similar tools a buying question.
TechCrunch reported that the company added roughly 300 customers a month after launch and reached $4 million in annual recurring revenue within ten months, evidence that investors don't expect this need to fade.
For a startup or SaaS team without a dedicated growth hire watching AI citations, the shift looks less exotic than a funding round.
It means writing comparison and pricing pages that state facts plainly enough for a model to lift them, and keeping product claims consistent across the site so an AI summary never has to guess which version is current.
Teams building this discipline into their own SaaS product often find it's the same habit that keeps onboarding copy accurate: one version of the truth, repeated everywhere.
None of this makes ordinary SEO worthless. Ranking well is still how a page gets read by the systems summarizing it, since most AI answers draw on the same organic results underneath.
What changes is the target: a page has to earn a citation inside someone else's answer now, not just a slot on a results page, and that calls for the kind of structured, fact-forward writing our AI and automation work is built around.
Start with the pages your buyers actually compare you on, and make sure the facts on them hold up regardless of who reads them next.
Cover photo by cottonbro studio on Pexels
Sources
- Do people click on links in Google AI summaries? — Pew Research Center
- Will traffic from search engines fall 25% by 2026? — Search Engine Land
