The model that connects search impressions to website traffic to pipeline is breaking. Not slowly, not theoretically: Ahrefs data from December 2025 shows AI Overviews now correlate with a 58% lower click-through rate for position-one rankings. That's up from 34.5% in April 2025. In eight months, the suppression nearly doubled.
If your demand gen forecast assumes stable conversion rates from organic search, you're forecasting from a baseline that no longer exists.
The Math Your Board Hasn't Seen
Let me walk through what this means for a typical B2B marketing org.
Current data puts Google's zero-click rate at 64.82%, up from roughly 50% in 2019. That trend predates generative AI, but AI Overviews accelerated it. When those overviews appear, the zero-click rate jumps to 83%. You're left with 17% of potential traffic instead of 40%.
Now layer in the rollout curve. Semrush data shared with Digiday shows AI Overviews appeared in about 20% of U.S. desktop searches by May 2025. That percentage has plateaued for now, but Google's AI Mode is still expanding, and the queries triggering overviews skew heavily toward informational intent: exactly the top-of-funnel content most B2B teams rely on for awareness and early-stage pipeline.
Here's the sensitivity table your CFO will want:
If 30% of your target keywords start showing AI Overviews, and those overviews suppress CTR by 58%, your organic traffic from those keywords drops by roughly 17%. If you're forecasting pipeline contribution from organic at last year's rates, you're overstating by that margin. If 50% of your keywords get hit, the gap widens to 29%.
This isn't a content quality problem. ZipTie.dev's analysis documents HubSpot seeing 70-80% traffic declines and DMG Media experiencing an 89% CTR drop on affected queries. These are well-resourced teams with strong SEO fundamentals. The structural shift doesn't care about your domain authority.
Why Traditional Measurement Now Lies to You
Your rank tracking tools still show positions. Google Search Console still shows impressions. But those metrics have decoupled from the outcome that matters: clicks that become pipeline.
One CMO described the disconnect: ranking #2 for a main keyword, 12,000 impressions, 340 clicks. That's a 2.8% CTR where she'd historically seen 8-9%. Her SEO rankings were stable. Her content hadn't changed. The problem was where users stopped: in the AI Overview, before they ever reached her site.
The forecasting implication is severe. If you're using historical CTR-to-MQL conversion rates, you're building on sand. The funnel's top has narrowed, but your model doesn't know it yet.
GoodFirms' 2026 survey found 43% of marketers now optimize for AI search, but only 14% measure it. That's an 86% blind-spot rate. Most teams are flying instruments that no longer reflect the terrain.
The Referral Quality Offset (and Its Limits)
There's a counterargument worth modeling. The clicks that survive AI Overviews convert 23% better than pre-overview traffic. Users who click through have already read a summary and want deeper information. They're higher intent.
Semrush and Seer Interactive data suggests AI search visitors convert 4.4x to 9x better than traditional organic visitors. That's meaningful. If your volume drops 50% but conversion rates triple, you might break even on pipeline contribution.
But here's the sensitivity risk: that conversion lift assumes your content is what users click through to see. If you're not cited in the AI Overview, you don't get the high-intent traffic. You get nothing.
Seer Interactive found brands cited in AI Overviews earn 35% more organic clicks and 91% more paid clicks than uncited competitors on the same queries. The gap between winners and losers is widening, not narrowing.
Reforecasting: A Two-Week Pilot
Before your next pipeline review, run this diagnostic:
Week one: Pull your top 100 pipeline-contributing keywords from the last four quarters. Check which ones now trigger AI Overviews (Ahrefs, Semrush, or SEOClarity all offer this filter). Segment your historical traffic and conversion data by AI Overview presence versus absence.
Week two: Build two forecast scenarios. Scenario A assumes current AI Overview penetration holds steady at 20% of queries. Scenario B assumes penetration doubles to 40% over the next 12 months (consistent with the March-to-May 2025 acceleration pattern). Apply the 58% CTR suppression factor to affected keywords. Compare both scenarios to your current forecast.
The delta between your current forecast and Scenario B is your exposure. If it's material, you need to brief your CFO before the board does the math themselves.
What Reallocation Looks Like
Google's May 2026 updates added in-line links, website previews, and "deep dive" sections to AI Overviews. These are attempts to restore some click-through, and early data suggests they help at the margins. But the structural incentive remains: Google wants users to stay on Google properties.
The strategic response isn't to abandon organic. It's to reallocate investment toward the queries and content types where clicks still flow, while building measurement for the visibility that doesn't click.
That means tracking AI citations as a KPI alongside rankings. It means investing in content formats that AI platforms cite (original research, proprietary data, expert commentary) rather than commodity explainers that get summarized and discarded. It means treating email, community, and owned audience as pipeline insurance against search volatility.
Gartner's forecast of a 25% decrease in traditional search traffic by 2026 appears to be playing out. The teams that adjusted their models six months ago are already operating with realistic baselines. The teams that didn't are about to explain a miss they didn't see coming.
Your forecast is only as good as its assumptions. The assumption that search traffic converts at historical rates is no longer defensible. Update the model, or the model will update your credibility for you.
