Why AI Visibility Influences Revenue More Than Your Analytics Show
ChatGPT sent one of our US e-commerce clients 5,416 sessions over the period we analysed. Google sent 82,401.
On volume, it’s no contest. But ChatGPT visitors converted at 7.18%, twice Google’s 3.59%, and each one was worth around $1.40 compared with 53 cents from Google.
That gap is why understanding how AI visibility influences revenue matters far more than your referral numbers suggest.
Tim Cameron-Kitchen, Founder of Exposure Ninja, uses our own client data and published research to explain why AI Search traffic performs so well, why most of its impact never appears in your analytics, and how to get your brand recommended in the conversations that lead to sales.
AI Traffic Is Smaller but Worth More
You’ll see claims everywhere that AI traffic converts better than Google traffic.
Tim Cameron-Kitchen points out the obvious contradiction: “Yes, traffic from AI tools like ChatGPT might be more valuable per visitor, but the amount of that traffic that it’s sending is so much less. So how can such a small traffic source be so valuable?”
For one of our eCommerce clients, Google drove 4,730 key events and around $44,000 in revenue. ChatGPT drove 633 key events and around $7,600.
Google wins on totals, but the growth story is very different.
Year-on-year, Google traffic grew 12% and Google revenue grew 40%. ChatGPT traffic grew 161%, and revenue from ChatGPT grew almost 400%.
“So yes, Google is contributing more visits and more conversions and more revenue, but ChatGPT is growing ultra-fast and actually the traffic is performing way better,” adds Tim Cameron-Kitchen.
This isn’t a one-off. Tim Cameron-Kitchen describes it as “a fairly common pattern that we see across lots of our clients.”
A lead generation client shows the same shape. AI platforms sent around 1% of the traffic that Google did, and drove 46 leads against organic search’s 1,700. Yet organic search converted at just under 6%, while AI traffic converted at just under 14%.
Your Analytics Only Show a Fraction of AI’s Influence
There’s a catch that makes AI referral data look even smaller than it really is.
A Similarweb clickstream study looked at what people do after an AI platform recommends a brand. 55% went on to search for that brand in Google. 20% went straight to the brand’s website.
Tim Cameron-Kitchen explains what that means for your reporting: “55% of these people are going to Google and searching for you, so they’re showing up in this organic bucket, and 20% of those people are going directly to your website. So you don’t even see that AI tool influence in your analytics platform.”
This means that the clicks you can see from ChatGPT are only part of the picture. Much of the value AI Search creates lands in your branded organic and direct traffic, where it gets credited to other channels.
For marketing leaders, that’s a real risk. Judge AI Search on referral traffic alone, and you’ll undervalue it.
Three Reasons AI Traffic Converts Better
1. Visitors Arrive Hyper-Qualified
On Google, someone might search “best accounting software”, then click through several sites to work out which one fits.
In ChatGPT, they’re more likely to ask something detailed, like which accounting software suits a UK business with 40 employees that needs multi-currency invoicing, integrates with Shopify, and costs less than £100 a month.
When ChatGPT answers with a specific recommendation, Tim Cameron-Kitchen notes that “it’s evaluated all of my criteria, so this solution matches much more closely what I’m looking for.”
The Google searcher is still researching. The ChatGPT user has already filtered their options.
2. Visitors are Further Down the Funnel
Traditional search journeys are long. Someone considering a hip replacement might run 10, 15 or 20 searches over weeks, clicking through websites without ever converting.
AI tools compress that journey. People can learn about a topic and get a tailored recommendation in a single conversation.
You may miss the early research clicks, but the visitor you do get is ready to act.
3. AI Recommendations Build Trust
Most people understand that ranking first in Google isn’t an endorsement. Tim Cameron-Kitchen puts it simply: “In search, people know that ranking number one doesn’t necessarily mean that Google is actually recommending you. It just means that your website is showing at the top of the list.”
An AI recommendation feels different because it’s based on the criteria the user has shared. A first-time buyer who describes their deposit, income and self-employed status, then gets a named mortgage broker back, sees that as a personalised answer.
Users still evaluate the recommendation for themselves. But they arrive with context and confidence that you’re a relevant option.
So, although we might not have got traffic from this user at various different stages of their journey, by the time we do get the click, they are really qualified, they understand the entire process, and they’re ready to move forward because they’ve identified that you’re a good fit for them.
Tim Cameron-Kitchen
Founder
How to Get Recommended in the Conversations That Drive Sales
Most brands need both: Google for volume, and AI Search for fast-growing, high-value traffic. In the video, Tim Cameron-Kitchen uses Semrush One, which tracks visibility across Google alongside ChatGPT, Google AI Overviews, Google AI Mode and Gemini.
Track Recommendations, Not Just Mentions
Being mentioned isn’t the same as being recommended. Tim Cameron-Kitchen explains the difference:
“It’s one thing to just be mentioned by an AI tool to someone who’s researching about a certain topic, but it’s much better to have that AI tool say, ‘You need to go with this brand because X, Y, Z.'”
Looking at Nuffield Health as an example, he shows how popular AI prompts around NHS and online GP services point to content opportunities. Building content around those topics improves the chances of being recommended for high-volume, commercially relevant prompts.
It’s also worth reviewing what AI tools say about your brand, positive, neutral or negative. Semrush’s AI Strategic Opportunities feature suggests questions to answer, and sometimes business improvements to make, so AI tools recommend you more often.
Find the Gaps Your Competitors Are Filling
Comparing your brand against a competitor shows which prompts and topics they’re visible for and you’re not. Not every gap will be relevant, but the ones that are point directly to content your website is missing.
Compare Your AI and SEO Visibility
Strong Google rankings don’t guarantee AI visibility. In the video, Nuffield Health scores 77 for AI visibility and 69 for SEO, performing well in both. L&C Mortgages has decent organic performance, but much weaker AI Search visibility.
The priority is to show up in bottom-of-funnel conversations, when the AI tool is about to recommend someone.
The Question to Ask Instead of “How Much AI Traffic Am I Getting?”
Our client data says AI traffic converts better, and significantly so. But Tim Cameron-Kitchen argues the bigger lesson is where buying decisions are now being made.
“Whereas previously people would’ve come to your website to establish if you’re a good fit for them, now from AI tools, people are coming to your website having already done that analysis and already done the research and comparisons with different options.”
So low AI referral numbers don’t mean you’re invisible in AI, or that AI isn’t driving revenue through other channels.
The better question, in Tim Cameron-Kitchen’s words, is: “Is my brand being found in the sort of conversations that lead to people spending money?”
Tracking as much AI referral traffic as you can is still important. Just don’t treat it as the full measure of how AI visibility influences revenue.
Want your brand recommended in the AI Search conversations that lead to sales? Our AI Search Optimisation specialists help mid-market and enterprise brands get found and recommended in ChatGPT, Google AI Overviews and beyond, then turn that visibility into qualified leads and revenue. Book a consultation call with the team at Exposure Ninja today.
