The signs AI search is sending you customers
A plumbing business can tell AI search is bringing in jobs when new callers mention asking ChatGPT, Gemini, or Siri before dialing, when website analytics show visits arriving from AI assistant referrer domains, and when the business's name and service area consistently show up when someone asks an AI engine to recommend a plumber nearby. None of these signs require guesswork; they show up in conversations with customers, in web traffic reports, and in repeated checks of what the engines actually say.
These signals matter because AI search engines don't work like a traditional search results page. Answer engine optimization (AEO), the practice of shaping content so AI tools can find and repeat it accurately, and generative engine optimization (GEO), the broader discipline of earning visibility inside AI-generated answers, both aim at the same outcome for a plumber: getting named when a homeowner asks an assistant "who fixes water heaters near me" instead of typing that same question into a search bar. Tracking whether that's happening takes a few habits, not a dashboard full of new tools.
Asking new callers how they found you
Every plumbing business already asks some version of "how did you hear about us?" That single question, asked consistently and logged consistently, is the fastest way to catch AI-driven jobs before any report or analytics platform confirms it. A caller who says "I asked ChatGPT for a plumber near me" or "Google's AI thing gave me your number" is direct evidence no traffic report can match.
The trick is training whoever answers the phone to listen for specific phrasing and write it down the same way every time. "Found you online" is too vague to act on. "Asked an AI chatbot" or "the AI answer at the top of Google" are specific enough to count separately from a plain Google search or a referral from a neighbor. Over a few months, a simple tally of these mentions, even a sticky note by the phone or a field in the booking software, shows whether AI-sourced calls are rising, flat, or barely happening yet. It also tells the owner which services get mentioned most, since customers often repeat back the exact problem they described to the AI tool, like "burst pipe" or "no hot water," which hints at what the AI's answer highlighted about the business.
Watching for AI-referred traffic
Website analytics can show whether visits are arriving from AI assistants, even though these platforms don't always label themselves the way a search engine referral does. Visits coming from domains associated with ChatGPT, Perplexity, Copilot, or Gemini, or from links embedded in an AI-generated summary, will appear in a site's traffic source reports under referral or direct traffic categories, depending on how the platform passes that information along.
Checking this regularly matters because it separates curiosity clicks from real interest. A visitor who lands on a plumbing site's emergency-service page after an AI assistant named the business for "24 hour plumber" is a stronger signal than a visitor who bounced off the homepage after ten seconds. Reviewing which pages AI-referred visitors land on and how long they stay shows whether the AI's answer sent someone who was actually ready to book, or someone still comparing options. Over time, a rising share of traffic from these sources, even a small one, indicates that AI engines are treating the business as a credible answer to nearby plumbing questions.
Tracking named mentions in engines over time
Watching how often a plumbing business gets named, by name, when someone asks an AI engine a relevant question is the most direct way to measure AI search visibility. This means periodically asking ChatGPT, Gemini, Perplexity, and Google's AI Overviews questions a real customer might ask, such as "who's a good emergency plumber in your city" or "which plumbing company installs tankless water heaters near me," and recording whether the business appears, in what position, and alongside which competitors.
This kind of tracking works best as a repeated exercise, not a one-time check, because AI-generated answers change as engines update their sources and as review counts, website content, and business listings shift. A plumbing owner who checks the same handful of questions every few weeks builds a record of whether their name is appearing more consistently, whether a competitor has started showing up instead, and whether the business's description in these answers is accurate. Small details matter here: an AI answer that lists the wrong service area or an outdated phone number is a visibility problem worth fixing even if the business is technically being mentioned.
A simple monthly review for a plumbing owner
A monthly review keeps all three signals, caller mentions, AI-referred traffic, and named mentions in engine answers, from becoming scattered notes nobody revisits. Set aside time once a month to total up how many callers mentioned an AI tool, check the analytics report for AI-referral traffic trends, and re-run the same set of test questions across ChatGPT, Gemini, and Perplexity to see if the business's name, service area, and details are still showing up accurately.
The value of doing this monthly, rather than sporadically, is pattern recognition. One caller mentioning an AI assistant could be a fluke. A pattern across ten callers over a month, paired with a steady trickle of AI-referred website visits and consistent naming in test questions, is a real trend worth paying attention to when deciding where to focus review generation, website updates, or local listing accuracy. A plumbing owner who tracks this monthly will notice a drop in mentions or a competitor creeping into answers long before it shows up as a dip in booked jobs.
When the AI answer names someone else instead
Picture a homeowner standing in a flooding basement at eleven at night, phone in hand, asking an AI assistant "who can fix a burst pipe near me right now." The assistant answers instantly, naming a plumbing company two towns over, describing their emergency response time, and offering a link to book. The homeowner, with water rising and no reason to second-guess the answer, calls that number instead of searching further.
That homeowner never saw a list of ten options to compare. They got one confident answer, and it wasn't the business that's been serving that neighborhood for years. This is the moment the monthly review is meant to catch before it happens again: not after the AI answer already sent the job down the street.