The average website gets 1.08% of its traffic from AI chatbots. That sounds tiny — roughly one visitor in a hundred. But here’s the kicker: that number comes from 13,770 domains across ten industries, and it’s growing about 1% month over month. ChatGPT alone drives 87.4% of those referrals. If you’re not measuring this, you’re flying blind on a channel that’s already reshaping how people discover brands.
Most site owners check their analytics, see “organic search” and “direct,” and call it a day. They miss the AI layer entirely. A potential customer asks ChatGPT for “best project management software for small teams,” your brand gets cited in the answer, they never click — but two weeks later they search your brand name directly and convert. Your analytics says “direct.” The reality? AI influenced that sale. The attribution gap is real, and it’s getting wider.
Conductor’s 2026 AEO/GEO Benchmarks Report dropped this data in August. It’s the first large-scale benchmark we’ve seen for AI referral traffic. The study covered May through September 2025 — so it’s already a few months old, and the channel has almost certainly grown since. But the 1.08% figure gives us something we’ve never had before: a baseline. Not a guess. A baseline.
The Number That Changes Everything: 1.08%
One percent. That’s the headline. Across 13,770 brand domains in ten industries, AI chatbots and AI Overviews delivered 1.08% of total traffic between May and September 2025. ChatGPT accounted for 87.4% of those referrals. The channel grew roughly 1% month over month during the study period.
Those are macro numbers. Your mileage will vary — a lot.
If you’re in IT, the benchmark jumps to 2.8%. Consumer Staples hits 1.9%. Communications Services and Utilities? Barely a blip. Publishers are a different story entirely — Conductor found AI sources account for less than 1% of publisher pageviews. The 1.08% figure blends brands and publishers across industries. It’s a directional reference, not a target for your specific site.
Here’s why that matters: if you run a B2B SaaS company and you’re getting 0.3% AI referral traffic, you might think you’re failing. You’re not. You’re below the cross-industry average but possibly above the benchmark for your category. Conversely, if you’re an IT consultancy getting 2%, you’re right on track — not crushing it.
The benchmark is useful because it finally lets you answer “is this normal?” with data instead of vibes.
AI Traffic Is Not Just Clicks — It’s Visibility
Here’s where most people get it wrong. They look at referral reports, see 1.08%, and think “AI isn’t a real channel yet.”
Wrong question.
A referral records a visit that arrives from a traceable source. It does NOT capture every way an AI answer influences a decision. Someone sees your brand cited in ChatGPT. They don’t click. Two days later they search your brand name directly. They convert. Analytics says “direct.” The AI touchpoint? Invisible.
This is the attribution gap. Conductor’s report calls it out explicitly: AI answers and AI Overviews may influence which companies, publications, or products users consider, even when conventional analytics attributes no visit to an AI source.
ChatGPT’s 87.4% share of AI referrals makes this even trickier. Broad AI referral figures are heavily influenced by one platform’s behavior. If ChatGPT changes how it cites sources — and it will — your referral numbers shift overnight. That’s not a reason to ignore the channel. It’s a reason to measure it differently.
You need two measurement layers:
1. Click data — AI-referred sessions in your analytics (trackable, but incomplete)
2. Citation visibility — how often your brand appears in AI answers (influential, but harder to track)
Most teams only do #1. They’re missing half the picture.
Industry Benchmarks: Your Mileage Will Vary
The cross-industry 1.08% is a blunt instrument. Conductor’s breakdown shows why:
| Industry | AI Referral Share |
|---|---|
| IT | 2.8% |
| Consumer Staples | 1.9% |
| Cross-industry average | 1.08% |
| Communications Services | <1% |
| Utilities | <1% |
Publishers need their own baseline entirely. The study examined brand domains across multiple industries. Publisher-specific reporting shows AI sources account for less than 1% of publisher pageviews. These findings coexist because they measure different populations, metrics, and traffic patterns.
If you’re a publisher, the 1.08% figure doesn’t apply to you. If you’re a B2B brand, it might understate your opportunity. The practical conclusion: AI discovery is rising and deserves attention, but it doesn’t currently dominate referral patterns or replace established channels. You need your own baselines for AI-referred sessions, branded demand, citation visibility, and the effects of indexing policies.
How to Actually Measure AI Impact (Step-by-Step)
You don’t need enterprise tools to start. Here’s the beginner-friendly setup:
Step 1: Create an AI referral channel group in GA4
Go to Admin → Data Settings → Channel Groups → Create new channel group. Name it “AI Referrals.” Add a rule: Source contains any of: chatgpt.com, claude.ai, gemini.google.com, perplexity.ai, you.com, phind.com, bing.com/chat, copilot.microsoft.com. Save. Apply to your reports.
Now you’ll see AI-referred sessions separated from organic, direct, and referral. It’s not perfect — some AI traffic comes through masked referrers — but it catches the majority.
Step 2: Set up a simple citation monitoring routine
Once a week, ask your top 10 strategic questions in ChatGPT, Claude, and Gemini. Questions your customers actually ask. “Best [your category] for [use case].” “Top [your category] tools 2026.” “[Your competitor] vs [your brand].” Note: are you cited? Is the information accurate? Does the AI link to you?
Track this in a spreadsheet. Date, platform, question, cited (Y/N), position (first/second/third mention), link included (Y/N). Over time you’ll see patterns. This is your citation visibility baseline.
Step 3: Connect AI signals to branded search lift
In Google Search Console, filter for queries containing your brand name. Look at impressions and clicks over the last 90 days. Correlate spikes with your citation monitoring. Did a week where ChatGPT cited you heavily correspond to a branded search lift? That’s your attribution bridge.
Step 4: Monitor AI Overview appearances
In Search Console, go to Search Results → Search Appearance → AI Overviews. This shows which of your pages appeared in AI Overviews, impressions, and clicks. It’s limited data — Google doesn’t show the actual AI Overview text — but it’s a start.
Build Your AI Visibility Governance Model
Conductor recommends a practical governance model. Here’s the beginner version:
| Weekly | Monthly | Quarterly |
|---|---|---|
| Check top 10 queries in ChatGPT/Claude/Gemini | Review GA4 AI referral channel report | Update strategic question list |
| Log citations in spreadsheet | Check Search Console AI Overviews | Correlate AI signals with branded search |
| Compare AI referral trends MoM | Adjust content strategy based on gaps |
Start with a Google Sheet. Columns: Date, Platform, Query, Cited (Y/N), Position, Link (Y/N), Notes. That’s it. No fancy tools required. The discipline matters more than the tool.
The goal isn’t to assign every conversion to AI. It’s to treat AI as one potential influence in a longer customer journey and document the limits of your attribution. When someone asks “where did this customer come from?” you can say “they searched our brand directly, but we were cited in ChatGPT for their category question two weeks prior.” That’s a defensible answer.
Takeaway: Don’t Wait for the Flood — Start Measuring Now
The 1.08% benchmark is a snapshot from mid-2025. The channel has grown since. ChatGPT’s dominance (87.4%) means one platform’s citation behavior drives most measurable referrals — but citation visibility across all platforms influences discovery in ways your analytics will never show.
You have two choices: ignore it until it’s too big to miss, or build a simple measurement habit now. The latter takes 30 minutes a week. The former leaves you scrambling when AI referral traffic hits 5% or 10% and you have zero historical data.
Set up the GA4 channel group today. Run your first citation check this week. Start the spreadsheet. In six months, you’ll have something most competitors don’t: a baseline, a trend line, and the ability to say “here’s how AI actually influences our business” with evidence instead of guesses.
The traffic is small. The influence is not. Measure what matters.
