AI Overviews Cut Clicks 38% — The WooCommerce Traffic You Cannot Attribute
A randomized field experiment found that Google AI Overviews causally reduce outbound organic clicks by 38% and push zero-click searches from 54% to 72% when they appear. AI Overviews trigger on approximately 42% of queries. For WooCommerce stores, this is not just a traffic problem — it is a measurement problem. The sessions that still arrive often come via branded search or direct URL after the user read an AI Overview, and GA4 attributes zero of them to the original organic query. Server-side first-party data capture and structured data for citation eligibility are the two durable responses.
The First Causal Proof
A randomized experiment isolated what observational studies could only suggest — AI Overviews cause the click reduction, not correlation.
On April 3, 2026, researchers Saharsh Agarwal and Ananya Sen published a working paper on SSRN titled “Google AI Overviews and Publisher Traffic: Evidence from a Field Experiment.” This was the first randomized field experiment measuring the causal impact of AI Overviews on search behaviour, and the results were unambiguous.
The researchers built a Chrome extension that randomly assigned 1,065 US participants to one of three groups. The control group saw Google Search normally. A “Hide AIO” group had the extension remove AI Overviews in real time. A third group was redirected to Google’s AI Mode for all searches. The study ran for two weeks per participant between January and February 2026.
AI Overviews appeared on 42% of queries. Removing them increased outbound clicks from 0.38 to 0.61 per search — a 38% causal reduction in clicks when AI Overviews were present. Zero-click searches rose from 54% to 72%. Over 95% of users in the Hide AIO group did not detect any changes during the study, and removing AI Overviews produced no measurable change in user satisfaction or ease of finding information.
Previous studies from Ahrefs, Seer Interactive, and BrightEdge had measured click reductions ranging from 34.5% to 61%, but those were observational — they compared queries with and without AI Overviews without controlling for the differences between those queries. The Agarwal and Sen study isolated the variable. The click reduction is caused by AI Overviews, not by the type of query that triggers them.
A randomized field experiment by Agarwal and Sen found AI Overviews causally reduce outbound organic clicks by 38% and push zero-click searches from 54% to 72% on queries where they appear.
You may be interested in: Half of Your Buyers Consulted AI Before They Arrived — Your GA4 Attributed Zero of Them
The Zero-Click Acceleration
The broader zero-click trend is accelerating faster than at any point in the last decade.
SparkToro’s 2026 research, based on Similarweb clickstream data, found that 64.82% of all Google searches now end without a click. That figure climbed from roughly 50% in 2019 to 60% in 2024 to nearly 65% in 2026. The decline from June 2025 to May 2026 alone was 8 percentage points — the fastest acceleration of zero-click behaviour in the last decade.
AI Overviews appear on more than 20% of all Google searches according to SparkToro’s methodology. When they do appear, click-through rates drop by nearly 60%. Google’s AI Mode — still in early deployment — produces a 93% zero-click rate according to Semrush, more than double the standard AI Overview figure. AI Mode had surpassed one billion monthly users by Google I/O 2026, with query volume more than doubling each quarter.
The trajectory is clear. Less than one third of Google searches now send a click to any website. For WooCommerce stores that built their acquisition model on organic search traffic, the input to the funnel is structurally shrinking — and the shrinkage is not temporary, it is the new architecture of search.
| Search State | Zero-Click Rate | Source |
|---|---|---|
| Standard Google (no AI Overview) | 54% | Agarwal & Sen 2026 |
| Google with AI Overview present | 72% | Agarwal & Sen 2026 |
| All Google searches (aggregate) | 64.82% | SparkToro / Similarweb 2026 |
| Google AI Mode | 93% | Semrush 2026 |
SparkToro’s 2026 research found that 64.82% of all Google searches now end without a click, with AI Overviews on 20% or more of searches reducing click-through rates by nearly 60%.
The Attribution Gap GA4 Cannot See
The sessions that still arrive are misattributed, and the influence that drove them is invisible to analytics.
The click reduction is the visible problem. The attribution gap is the hidden one. When a user searches for “best WooCommerce tracking solution,” reads the AI Overview that synthesises information from several sources including your site, and then later types your brand name directly into the browser or searches for it by name — that session arrives in GA4 as direct traffic or branded organic. The original informational query that surfaced your store to the user is never connected to the eventual visit.
This is not a bug in GA4. It is a structural limitation of browser-based analytics. GA4 sees the session that starts on your site. It does not see the search that happened before the user decided to visit. It does not know the user read an AI Overview that cited your product. It cannot attribute the branded search to the informational query that created the brand awareness.
The result is a growing category of “dark traffic” — sessions that are commercially valuable, driven by discoverable content, but invisible to the attribution model that decides where you spend your marketing budget. Every WooCommerce store owner who looks at their GA4 and sees rising direct traffic with declining organic may be looking at the AI Overview effect without recognising it.
AI-driven visitors convert at 4.4 times the rate of standard organic visitors according to Semrush data. The visitors arriving through AI-influenced paths are fewer in number but significantly higher in intent because they have already read a summary, compared alternatives, and decided to investigate further. If your measurement system cannot see them, your budget allocation will underinvest in the content and structured data that generated them.
Where Ecommerce Sits in the Impact Spectrum
Transactional queries are less affected than informational ones — but informational queries drive the top of the funnel.
The AI Overview impact is not uniform across query types. Ecommerce shopping queries trigger AI Overviews on only 3.2% of searches, down from 29% initially. Google scaled back AI Overviews on transactional queries because AI-generated responses were not converting into sales — users still need to visit a store to buy. That pullback creates a temporary safe zone for product-page and category-page traffic.
The vulnerability is in the informational layer. Product comparisons, buying guides, how-to content, category education — the queries that build awareness and shape purchase intent before the transaction — are heavily affected. Health-related queries trigger AI Overviews on 60.7% of searches. Technology queries are similarly impacted. These are the informational queries that WooCommerce stores use to attract top-of-funnel traffic, build email lists, and create the brand recognition that eventually converts into sales.
When informational traffic declines, the pipeline that feeds transactional traffic narrows. A WooCommerce store might see steady product-page sessions while the blog content that originally introduced customers to the brand quietly loses its organic reach. The effect shows up months later as a decline in branded search and returning visitors — by which time the causal connection is difficult to trace without the right data.
AI-driven visitors convert at 4.4 times the rate of standard organic visitors according to Semrush data, meaning the traffic that does arrive from AI-influenced search is higher-value than what GA4 labels as direct.
You may be interested in: AI Overviews Hit 48% of Searches — WooCommerce CTR Impact
The Measurement Response
Two parallel tracks — earning AI citations and capturing first-party data — form the durable response.
The first track is structured data. Pages with comprehensive schema markup are cited in AI Overviews at higher rates. Product schema, FAQ schema, review schema, and breadcrumb schema help Google’s AI systems identify your content as a reliable source worth citing. Being cited in an AI Overview does not produce a click, but it produces brand exposure at the exact moment the user is researching a purchase. The brand recognition compounds over time and surfaces as the branded search and direct traffic that arrives later.
The second track is first-party data capture at the server level. When the user who read your AI Overview citation eventually arrives — via branded search, direct URL, or a referral link — a server-side data pipeline captures the full session at the WordPress hook level. The capture includes referrer data, landing page, UTM parameters if present, and customer identity once the user engages with the checkout. This data goes into your own warehouse, not GA4’s sampled interface with its 14-month retention ceiling.
AI referral traffic currently accounts for 1.08% of all website traffic according to Conductor’s 2026 benchmarks, growing approximately 1% month-over-month. That number is small now. But it is growing, and the visitors it represents convert at multiples of standard organic. Capturing and attributing this traffic requires measurement infrastructure that exists independently of Google’s analytics — because the very platform reducing your clicks is also the one controlling the analytics that measure them.
For WooCommerce stores running the Transmute Engine™, this data flows directly into BigQuery alongside every other event the store generates. The AI-influenced session that GA4 classifies as direct sits next to the ad-driven session and the organic session, all captured at the same server-side source of truth. Budget decisions come from owned data, not from a platform whose product changes are the reason the data is incomplete.
You may be interested in: The Cumulative Tracking Loss Stack Hides Up to 50% of Your WooCommerce Customers
Key Takeaways
- AI Overviews causally reduce organic clicks by 38%: The Agarwal and Sen randomized field experiment is the first causal proof, not a correlation. Zero-click searches rise from 54% to 72% when an AI Overview appears.
- Nearly two thirds of Google searches now end without a click: SparkToro measured 64.82% zero-click in 2026, up 8 percentage points in a single year — the fastest acceleration in a decade.
- The sessions that still arrive are misattributed: Users who read an AI Overview and later visit via branded search or direct URL are invisible to GA4’s attribution model. The influence is real but the credit goes elsewhere.
- Ecommerce product queries are partially shielded: Only 3.2% of shopping queries trigger AI Overviews after Google’s pullback. But the informational queries that drive top-of-funnel awareness are heavily impacted.
- The measurement response is two-track: Structured data earns citations in AI Overviews for brand exposure. Server-side first-party data capture measures and attributes the sessions that arrive through AI-influenced paths.
A randomized field experiment by Agarwal and Sen found that AI Overviews causally reduce outbound organic clicks by 38% on queries where they appear. The same study found zero-click searches rise from 54% to 72% when an AI Overview is present. These figures represent a causal effect measured through a controlled experiment, not an observational correlation.
WooCommerce stores see two effects. First, fewer organic sessions arrive because users get answers without clicking. Second, sessions that do arrive after a user reads an AI Overview often come via branded search or direct URL, so GA4 attributes them to brand or direct traffic rather than the organic query that originally surfaced the store. The influence is real but the attribution is invisible.
Stores cannot reverse the zero-click trend, but they can respond on two fronts. First, structured data and schema markup increase the likelihood of being cited in AI Overviews, which drives brand recognition even without a click. Second, server-side first-party data capture ensures that sessions arriving via branded search or direct URL after an AI Overview exposure are still measured, attributed, and stored in the store’s own data warehouse.
Transactional ecommerce queries are less affected than informational ones. Ahrefs data shows ecommerce shopping queries trigger AI Overviews on only 3.2% of searches, down from 29% initially. Google scaled back because AI responses were not converting into sales. However, the informational queries that drive top-of-funnel discovery — product comparisons, how-to guides, category education — are heavily affected.
References
- Agarwal & Sen — Google AI Overviews and Publisher Traffic: Evidence from a Field Experiment (SSRN, April 3, 2026, revised June 17, 2026)
- Search Engine Journal — Study Confirms Google AI Overviews Cut Organic Clicks 38% (April 29, 2026)
- SparkToro / Similarweb — In 2026, Less Than One Third of Google Searches Still Send a Click (June 18, 2026)
- Digital Applied — Zero-Click Search Statistics 2026: Complete Data Guide (April 5, 2026)
- Semrush — AI Mode zero-click rate at 93%; AI-driven visitors convert at 4.4x standard organic (2026)
- Conductor — AI referral traffic at 1.08% of all website traffic, growing ~1% month-over-month (2026)
- SeoProfy — AI Overviews appear in ~20% of Google searches, reaching 37.2% in some countries (May 2026)
- Ahrefs via Pillitteri — Ecommerce shopping queries trigger AI Overviews on only 3.2% of searches (2026)
- Search Engine Land — Google I/O 2026: AI Mode surpassed 1 billion monthly users (June 2026)
If your WooCommerce store’s organic traffic is declining while direct traffic rises, you may be looking at the AI Overview effect without recognising it. See how Seresa’s server-side pipeline captures what GA4 misattributes.