What percentage of your traffic comes from AI search? Share your data

B2B SaaS here. Let me share our data:

Our numbers (200K monthly sessions):

• AI referrals: 4.7%
• ChatGPT: 2.8%
• Perplexity: 1.4%
• Other AI: 0.5%
• YoY growth: +520%

What drives our higher percentage:

We’re in a category people research heavily before buying. Complex B2B decisions = more AI research queries.

Quality observation:

AI traffic converts at 3.2x the rate of organic. These visitors are further in their journey and often specifically directed to us.

My theory on your 2.3%:

Depends on your category. 2.3% seems reasonable for many industries. Higher-consideration products/services tend to have higher AI traffic.

E-commerce perspective (general consumer goods):

Our numbers (1.2M monthly sessions):

• AI referrals: 0.8%
• Much lower than B2B

Why we think it’s lower:

For commodity products, people don’t ask AI - they just search and buy. AI traffic skews toward research, not transactions.

Where AI does appear:

“Best X for Y” category research. When we optimize product guides and comparisons, AI traffic increases.

The breakdown:

• Product pages: 0.3% AI traffic
• Guide/comparison content: 4.2% AI traffic
• Category pages: 0.9% AI traffic

AI traffic is content-type dependent, not just site-level.

Tech content publisher:

Our numbers (3M monthly sessions):

• AI referrals: 6.8%
• Growing at 400% YoY

We’re higher because our content is what AI recommends when people ask tech questions.

Quality metrics:

• AI traffic engagement: 4:32 avg time on page
• Organic engagement: 2:15 avg time on page
• Subscription conversion from AI: 2.1%
• Subscription conversion from organic: 0.7%

AI sends us better users. Worth optimizing for even if volume is still small.

Local service business (home services):

Our numbers (15K monthly sessions):

• AI referrals: 0.4%
• Barely measurable

Why so low:

People don’t ask AI for local services. They search “plumber near me” on Google/Maps.

Where AI does appear:

Content about how to evaluate/choose services. “What to look for in a roofer” type pages.

My conclusion:

AI traffic relevance depends heavily on industry. Local businesses shouldn’t obsess over AI - traditional local SEO still matters way more.

Aggregate data from multiple clients:

Average AI traffic by industry:

Key observations:

1. Research-heavy industries have higher AI traffic
2. Transactional/local industries have lower
3. Growth is strong across all categories
4. Quality (conversion) often beats quantity

Your 2.3% is right in the middle - normal for a general B2B context.

Analytics tracking perspective:

Warning: You’re probably undercounting.

AI traffic attribution is messy:

1. Some AI traffic comes with no referrer (appears as “direct”)
2. ChatGPT’s browse feature doesn’t always pass referrer
3. Users may copy/paste URLs (no referrer)

Better tracking:

• Create specific landing pages for AI contexts
• Use UTM parameters where possible
• Compare branded search lift (AI users often Google you after)
• Track overall patterns, not just direct referrals

Your 2.3% might actually be 3-4% if you could capture everything.

Amazing data from everyone. Here’s my synthesis:

Benchmark ranges:

• Tech/Publishing: 4-8% (high research)
• B2B SaaS: 3-6% (high consideration)
• General B2B: 2-4%
• E-commerce: 0.5-2%
• Local services: 0.2-1%

Our 2.3% in context:

Reasonable for general B2B. Room to grow, especially for research-oriented content.

Key insights:

1. Quality > quantity for AI traffic
2. Content type matters more than site average
3. We’re probably undercounting
4. Growth trajectory is what matters most

What I’m doing next:

1. Create AI traffic segment in GA4
2. Track conversion by traffic source more granularly
3. Identify which content types drive AI traffic
4. Optimize high-AI-traffic pages for conversion
5. Use Am I Cited to understand citation patterns

Thanks for all the benchmarks. Really helpful for setting expectations with leadership.

Projection for context:

Current state (early 2026):

• Average site: 1-5% AI traffic
• Growing at 300-500% YoY

Trajectory suggests:

• End of 2026: 3-10% AI traffic average
• End of 2027: 8-20% AI traffic average

The takeaway:

Current numbers seem small but growth is exponential. What’s 2% today could be 10% in 18 months.

Build the tracking and optimization infrastructure now. The volume will come.

Revenue perspective:

Don’t just track traffic - track pipeline.

For B2B, we track:

• AI traffic to MQL rate
• AI-influenced deals (touched AI content)
• Revenue attribution

What we found:

AI is 4% of traffic but 8% of pipeline. The quality difference makes it punch above its weight.

Measure what matters to the business, not just sessions.