Is AI Sending You Customers? How to Find Out
A practical self-audit to see whether AI answers actually drive business your way — ask the engines your customers' questions, check whether and how you're named, look for AI referral traffic in your analytics, and track it over time.
The fastest way to find out whether AI is sending you customers is to check directly: ask the engines the questions your customers ask, then look in your analytics for traffic they send you. If you're not named and no referrals show up, AI is quietly routing those customers to someone else.
Quick answer
You don't have to guess. Run a two-part self-audit — ask the assistants your real customer questions and note who they name, then check your analytics for referral traffic from AI tools. Do it monthly and you'll know exactly whether AI is a source of customers or a leak sending them to competitors.
How do I check whether AI is sending me customers?
Run the audit below. It takes under an hour and needs nothing but the assistants themselves and your analytics login.
- 1
List your real customer questions
Write down the exact things prospects ask before they hire — 'best [trade] in [city],' 'who can fix [problem] fast,' 'should I hire X or Y.' Use their words, not yours.
- 2
Ask every major engine
Put each question to ChatGPT, Google AI Mode, Perplexity, and Gemini. Note whether you're named, who else is, and whether your details are right.
- 3
Open your analytics
In your referral or traffic-source report, look for visits from chatgpt.com, perplexity.ai, gemini.google.com, and similar AI domains.
- 4
Log it and repeat monthly
Record who got named and how many referrals arrived. Run the same questions next month to see which way the trend is moving.
What am I actually looking for when I ask the engines?
Three things. First, are you named at all when you ask a buying question? If the assistant lists a couple of businesses and yours isn't among them, that's the headline result. Second, who is named instead — those are the competitors AI is handing your customers to. Third, is the description accurate — sometimes you're mentioned but with stale hours, the wrong services, or an old phone number, which quietly costs you the call. This is the same manual check behind how customers use AI to choose a business, run on your own name.
Where does AI referral traffic show up in my analytics?
In the referral or source section of whatever analytics you use. Look for domains like chatgpt.com, perplexity.ai, and gemini.google.com. It tends to come in small numbers but with unusually high intent — these are people who already got a recommendation and clicked through to act. One caveat: a lot of AI influence is invisible. A customer reads the answer, then searches your business name directly or just calls, leaving no AI referrer at all. So the referral count is a floor on AI's impact, not the whole of it. For the deeper version of this, see how to track AI citations and the concept of AI referral traffic.
What does it mean if the audit comes back empty?
It means AI isn't sending you customers yet — and that's a fixable finding, not a verdict. An empty result almost always traces to one of a few gates: an AI crawler can't read your site, your key page buries the answer instead of leading with it, or your off-site reputation is too thin for engines to trust you over a competitor. Diagnosing which gate is closed is the whole game — start with why isn't my site cited by AI.
How do I turn this into ongoing measurement?
Make it a habit, not a one-time scare. Keep a simple log — question, date, who was named, referrals that month — and re-run it on the same day each month.
Your monthly AI-customer audit
0 / 5
Each unchecked box is a place a competitor can beat you to the AI answer.
Over a few months that log tells you whether your visibility is rising, flat, or slipping — and that's the signal that tells you if your Answer Engine Optimization work is paying off.
The bottom line
You can find out today whether AI is a source of customers or a silent leak — the audit is free and takes an hour. If it comes back empty and you want the customers AI is currently sending elsewhere, book a call and we'll show you exactly which gate is closed and how to open it.
Related questions
You use AI every day — but is AI recommending your business?
The core divide between using AI and being found by it.
Read the full answer →How do I track AI citations?
A repeatable method for monitoring whether and how AI names you.
Read the full answer →Why isn't my site being cited by AI?
Usually one broken gate — access, a buried answer, or weak off-site trust.
Read the full answer →How do customers use AI to choose a business?
The new buying journey, from question to a shortlist of one or two names.
Read the full answer →What is AEO?
Answer Engine Optimization — being the source AI cites and recommends.
Read the full answer →What does it cost to be invisible in AI search?
The silent price of the customers you never see choose a competitor.
Read the full answer →Frequently asked questions
- How do I know if AI is sending me customers?
- Run a two-part check. First, ask ChatGPT, Google AI Mode, Perplexity, and Gemini the exact questions your customers ask and note whether you're named. Second, look in your analytics for referral traffic from AI tools. If you're named and seeing referrals, AI is working for you — if not, it's sending those customers elsewhere.
- Where does AI referral traffic show up in analytics?
- In your website analytics, check the referral or source report for domains like chatgpt.com, perplexity.ai, and gemini.google.com. It usually arrives in small but high-intent numbers. Some visits leave no referrer at all because the customer read the answer and then searched your name directly, so treat this as a floor, not a full count.
- What if AI names a competitor instead of me?
- That's the gap, and it's fixable. It means the engines found a clearer, more trusted answer elsewhere when a customer asked. The fix is to make your key page the best answer to that question and to strengthen the reviews and mentions engines trust — the discipline called Answer Engine Optimization.
- How often should I check whether AI recommends me?
- Re-run the same questions monthly and log who gets named each time. AI answers shift as engines update and as your presence changes, so a single check is a snapshot. Tracking it over time turns a one-off into a trend you can actually manage.