Two out of three Google searches in the U.S. now end without a click. That's 68.01% as of early 2026, up from 60.45% in 2024. A 7.56-point swing in two years.
That shift is a revenue measurement error, and it compounds every month your finance team evaluates content spend against click-derived pipeline. If your per-article economics model still assumes that ranking = traffic = revenue, you're systematically undercounting what your content is worth while simultaneously missing where the value is actually flowing.
We've spent the last quarter rebuilding the per-article economics model we use internally. The math surprised us, and we think it'll surprise most two-person marketing teams still running 2023-era attribution.
The Old Model Broke Quietly
Most B2B content teams calculate article ROI with a formula that hasn't changed in a decade. Rank position feeds estimated traffic, traffic feeds conversion rate, conversion feeds pipeline value, pipeline feeds ROI. Simple, clean, wrong.
The model assumes that Google SERPs function the way they did in 2021. They do not. AI Overviews now appear on 48% of searches, and that number peaked at 83% in the education sector. When an AI Overview is present, Ahrefs data shows a 58% reduction in CTR for the top-ranking result. Seer Interactive's study across 25 million impressions found organic CTR dropping from 1.62% to 0.61%.
So your position-one article didn't get worse. The SERP around it changed, and your measurement framework didn't notice.
Here's what makes this particularly painful for small teams: more than half (56%) of B2B marketers still struggle with attribution, even as 49% report that content directly impacts revenue. They know content works. They just can't prove it accurately anymore.
Building a Revised Per-Article Economics Model
We rebuilt the model around three variables that traditional frameworks ignore. Each one represents revenue your current spreadsheet is leaking.
Variable 1: Effective CTR, not position-estimated CTR. The old model looks at your ranking position and pulls an estimated CTR from a curve (position 1 = ~27%, position 2 = ~15%, etc.). But those curves were built before AI Overviews existed. The revised model applies a discount factor based on AIO presence probability in your target keyword cluster. For a B2B SaaS company targeting mid-funnel terms, we've seen AIO appearance rates between 35% and 55%. Apply the 58% CTR reduction to that proportion of your keyword portfolio, blend it with non-AIO CTR for the remainder, and you get an effective CTR that's 20-35% lower than what most teams report.
Variable 2: Citation-driven branded search value. This is the revenue stream nobody tracks. Brands cited inside AI Overviews get 35% higher organic CTR and 91% higher paid CTR compared to non-cited brands. Some niche publishers are growing revenue even as traffic declines because AI Overview citations function like word-of-mouth at Google scale. Users see your brand, don't click, but search for you directly later. That branded search lift never appears in your article-level attribution. It shows up as "direct" or "branded organic" traffic in GA4, completely disconnected from the content that triggered it.
Variable 3: True production cost allocation. Most content teams dramatically undercount production costs because they track freelancer fees and tool subscriptions while ignoring internal labor, promotion time, and ongoing maintenance. A "free" blog post written by your head of marketing still costs $400-800 in opportunity cost when you account for the 4-6 hours of research, writing, editing, and publishing. And that's before distribution.
Running the Numbers: A Concrete Example
Take a B2B SaaS company publishing 12 articles per month. Traditional model assumptions versus the revised model, side by side.
The traditional approach would estimate: average position 4 across target keywords, 8.1% CTR from the standard curve, 2,000 organic visits per article per year, 2.5% conversion rate, $150 average deal value at the top of funnel. That gives you $7,500 in attributed pipeline per article per year. Against a fully loaded cost of $600 per article (including internal time), you get a 12.5x return. Looks great on a slide deck.
Now run the revised version. Same position 4, but apply AIO discount: 45% of those keywords now trigger AI Overviews, each reducing CTR by 58%. Blended effective CTR drops to 5.2%. Organic visits fall to 1,284 per article per year. Same 2.5% conversion rate, same $150 deal value. Click-derived pipeline per article drops to $4,815. But we haven't added Variable 2 yet.
If your content gets cited in AI Overviews (and not all of it will; we'll address that), branded search lift contributes an estimated 15-25% additional pipeline attribution. Call it 20%. That adds $963 back. And if you're honest about production costs, including 5 hours of internal time at $75/hour, your true cost per article is $975, not $600.
Revised ROI: ($4,815 + $963) / $975 = 5.9x return.
Still positive. Still worth doing. But 53% lower than what the old model told your CFO.
The Reallocation Threshold
Here's where the math gets interesting. There's a specific point where traditional click-volume SEO produces a lower expected return per dollar than a hybrid strategy targeting both ranked clicks and AI citation surfaces.
We found that threshold by modeling two scenarios across a 12-month horizon for a team publishing 12 posts per month.
In Scenario A (pure ranked-click SEO), all 12 posts target traditional keyword clusters optimized for position and click volume. Effective CTR continues declining at the projected 15-25% annual rate. Monthly attributed pipeline: declining curve starting at roughly $57,780 and eroding ~2% per month as AIO coverage expands.
In Scenario B (hybrid visibility), 8 posts target ranked clicks and 4 posts are specifically structured for AI citation (question-format headers, concise authoritative answers, structured data, primary source positioning). The 4 citation-optimized posts generate 40% fewer direct clicks but produce 3x the branded search lift. Monthly attributed pipeline: flatter curve starting at $52,200 but holding steady or growing as AIO coverage expands.
Scenario A starts higher. Scenario B overtakes it around month 7.
That's the reallocation threshold. After 7 months, the hybrid model outperforms because its citation-derived pipeline compounds while the pure click model erodes. And the crossover happens faster if your vertical has above-average AIO penetration (SaaS, education, health, and finance all exceed 50%).
What Actually Changes in Your Measurement Stack
Measuring content marketing ROI for B2B SaaS comes down to four things most teams miss: conversion-path traffic, content-influenced pipeline, time-to-close comparisons, and AI citation share. We'd add a fifth: branded search lift segmented by content cohort.
The practical implementation requires three changes to your GA4 setup and one change to your editorial planning process.
First, extend your attribution lookback window to 90 days. The average B2B buyer journey spans 6-12 months, but identifiable content touchpoints cluster in the 60-90 days before conversion. GA4's default 30-day window misses half the story.
Second, create a branded search segment that isolates brand-name queries and correlate volume changes against your content publishing calendar. A spike in branded searches 2-3 weeks after publishing a piece that gets cited in AI Overviews is your citation ROI, even if GA4 attributes those sessions to "organic branded."
Third, track AI citation share manually. Yes, manually. There's no reliable automated tool for this yet (and if someone sells you one, be skeptical). Run your target queries through Google with AIO enabled, note which of your pages get cited, and log it weekly. Fifteen minutes of work that gives you data nobody else on your team is collecting.
The editorial change is more significant. 67% of SEO teams have already changed how they measure success since AI Overviews launched. The ones seeing results are shifting a portion of their editorial calendar from generic keyword-targeted posts toward what we'd call "citable assets": content structured to be the definitive, concise answer to a specific question, with original data or unique analysis that AI systems prefer to cite.
The Part Nobody Wants to Hear
Some of this is genuinely messy. Attribution for citation-driven revenue is imprecise. Branded search lift is a proxy, not a proof. And the 7-month crossover point we calculated depends on assumptions about AIO expansion rates that could shift if Google changes course.
But here's what we know with confidence: B2B content teams that measure ROI purely on click-derived pipeline are making budget decisions based on a number that's 30-50% too high. That doesn't mean content marketing is broken. B2B SEO still delivers an average 748% ROI over a multi-year horizon. The asset appreciates. The economics still work.
They just work differently than your spreadsheet says. And the teams that fix their measurement first will reallocate before their competitors even notice the leak.
References
-
Search Engine Land. "Google zero-click searches reach 68% in early 2026: Study." https://searchengineland.com/google-zero-click-searches-2026-study-479717
-
Digital Applied. "60% Zero-Click Searches: The 2026 SEO Crisis Strategy." https://www.digitalapplied.com/blog/60-percent-searches-zero-click-crisis-2026-seo-strategy
-
MediaNama. "Google AI Overviews Reduce Clicks By 58%, Study Finds." https://www.medianama.com/2026/02/223-google-ai-overviews-click-through-rates-58-study/
-
Digital Applied. "Content Marketing ROI 2026: Measurement Framework." https://www.digitalapplied.com/blog/content-marketing-roi-2026-measurement-framework
-
Positive Equator. "How To Measure Content Marketing ROI for B2B SaaS." https://positiveequator.com/how-to-measure-content-marketing-roi-for-b2b-saas/
