AI-referred traffic is now measurably more valuable than ordinary search traffic, according to Adobe Digital Insights' Q3 AI Traffic Trends Report (June 2026): in May 2026, AI-referred retail visitors converted at a 54% higher rate and generated 53% more revenue per visit than non-AI traffic — a complete reversal from a year earlier. This page collects the report's headline figures, each tied to its source and date.
Last reviewed: 17 June 2026. All figures below are from Adobe Digital Insights' Q3 AI Traffic Trends Report (June 2026) or the March 2026 Adobe Consumer Survey of 5,000+ US respondents, as noted. Adobe's web data covers more than one trillion visits to US retail sites and 100M+ SKUs — large, but single-vendor and self-reported, so read the direction as firmer than any one percentage, and hedge anything you quote as "as of mid-2026."
How fast is AI traffic growing in 2026?
AI-driven visit share kept climbing across every industry Adobe measured, well outside holiday peaks:
| Industry | AI visit-share growth (YoY, May 2026) | Source |
|---|---|---|
| Travel | +194% | Adobe Digital Insights, Q3 2026 |
| Retail | +138% | Adobe Digital Insights, Q3 2026 |
| Financial Services | +105% (19 straight months of growth) | Adobe Digital Insights, Q3 2026 |
Adobe's read: because AI visit share in May surpassed every month of 2025 without a seasonal lift, the shift is structural, not a holiday phenomenon. The March 2026 Adobe Consumer Survey adds that 54% of consumers say they are turning to AI more, and 58% had used AI in the past week. These are growth rates off a still-small base — large multiples, modest absolute share — which is the same pattern other trackers report (see the broader AI search statistics reference).
Does AI-referred traffic convert and spend more?
Yes — and the gap has inverted in the space of a year. For US retail in May 2026, Adobe reported AI-referred visitors outperforming non-AI traffic on every commercial metric:
| Metric (Retail, May 2026) | AI vs non-AI | Note |
|---|---|---|
| Conversion rate | +54% higher | A year earlier, AI converted at ~half the rate |
| Revenue per visit | +53% higher | Up from +37% in March; a year prior non-AI was worth 128% more |
| Engagement rate | +15% higher | Strongest advantage since tracking began (Oct 2024) |
| Time on site | +53% more | April peak was +55% |
| Pages per visit | +23% more | Up from a 13% gap two months earlier |
| Bounce rate | −36% (less likely to bounce) | AI held an 17–20% band vs ~27% for non-AI |
The one-line summary Adobe draws: AI assistants are sending shoppers who arrive with intent rather than a fleeting click. This is also why AI's commercial impact is easy to under-count in analytics — much of it shows up as direct traffic rather than a labelled AI referral, so it needs deliberate measurement.
What does the consumer survey say about AI shopping behaviour?
The March 2026 Adobe Consumer Survey (5,000+ US respondents) puts behaviour behind the traffic numbers:
- 39% have used AI assistants for online shopping; 85% of those say AI improved their shopping experience.
- 55% turn to AI for inspiration and ideas — most often before they begin shopping.
- 50% click the links an AI assistant provides when shopping; 27% complete purchases directly through those links.
- 79% feel more confident in a purchase after using an AI assistant; 69% say they are less likely to return an item bought with AI help.
- On trust: 66% agree GenAI provides accurate results, and 38% trust AI more than they used to.
The takeaway for brands: discovery increasingly starts in an AI answer, and the click — when it comes — carries higher intent. Being present and accurately described at the inspiration stage is what feeds the rest of the funnel, which is the core argument for getting recommended inside the answer.
Is the pattern the same in travel and financial services?
Broadly yes, with sector-specific detail. AI-referred visitors engaged more deeply across all three industries Adobe tracked:
| Sector (May 2026) | Engagement | Time on site | Bounce | Notable |
|---|---|---|---|---|
| Travel | +21% | +70% (up from 61% in March) | −41% | Conversion gap narrowed from ~86% (Oct 2024) to 28% (May 2026), 67% narrower YoY |
| Financial Services | +8% | +27% | −18% | 89% trust AI for financial recommendations without human input; 46% of those fully follow the advice |
For travel, Adobe reported 86% of travelers had an improved experience planning via an AI assistant, using it for research (48%), inspiration (44%), budgeting (30%), and packing (21%). For financial services, 24% of consumers now use AI assistants for financial needs. Across all three sectors the same behavioural signal holds: AI sends visitors who have already researched and arrive ready to act.
Which page types are most "AI-readable"?
Adobe's report introduces a citation readability score — its measure of how well a page's content can be parsed, understood, and surfaced by AI systems (structured, complete, and aligned to how models process information). Comparing the top 20% of companies by AI visit share against the bottom, Adobe found the biggest gaps at the entry and discovery layer:
| Page type | Readability advantage (top vs bottom performers) |
|---|---|
| Homepage | +52% |
| Search experience | +32% |
| Content pages (blogs, buying guides) | +23–30% |
| Exploration pages (brand landing, product detail) | +14–15% |
| Category / collection pages | +5% |
By retail sub-industry, citation readability (May 2026) ran: Cosmetics 63%, Electronics 56%, Sporting Goods 51%, Apparel 51%, Grocery 48%, Furniture & Home 47% — a 16-point spread Adobe attributes directly to content strategy, with editorial, ingredient-education, and spec-rich pages reading best to AI. This is the same lesson the corpus reaches from first principles: accessibility and clean structure predict AI-parseability, and structured, complete content is what gets cited over a near-identical competitor.
The numbers will move, but the direction is the story: AI sends fewer visitors than search today, yet each one is worth more — more likely to convert, spend, and stay — and the pages AI can cleanly parse are the ones that win the visit in the first place.
How should you use these figures?
Treat this as a dated reference, not a guarantee. Every number here is single-vendor and self-reported, drawn from Adobe's own web data and consumer survey; cite Adobe and the date when you reuse one, and prefer your own measured trend over any headline percentage. Because live retrieval favours recently-updated pages, we keep references like this on a refresh cadence and update figures substantively as new data lands.
The figure that ultimately matters is your own: whether AI engines surface, cite, and recommend your brand — and whether that AI-referred traffic converts for you — tracked over time across every engine. That measurement is exactly what Buffy Intel is built to provide.
