How We Measure Whether AI Search Is Actually Working For You
AI search — ChatGPT, Google's AI answers, Perplexity, and the rest — is where a growing share of New River Valley buyers now ask "who's the best" question before they ever hit a map or a website. The hard part is not getting mentioned. It is proving the work is paying off without fooling yourself with numbers that look good and mean nothing. Here is how we actually measure it, and what we deliberately ignore.
What we track first is inclusion: when someone asks an AI assistant for your kind of business in Christiansburg or the New River Valley, does your name come up, and where in the answer? We run the real prompts a local customer would type — your trade plus the town, plus "near Blacksburg," plus the specific problems people describe in their own words — and we log whether you appear, whether you appear first, and whether the AI describes you accurately. That last part matters: being mentioned with the wrong service area or a stale detail is a problem, not a win.
Second, we track the shape of what the AI says about you. Does it cite your strengths, your actual services, your real coverage area? AI systems assemble that description from your site, your reviews, and structured data. When we improve those inputs, the description sharpens — and we can show you the before and after in plain language, not a score.
Third, and most important, we track whether it turns into contact. AI referrals are notoriously hard to attribute because the click often does not carry an obvious tag, so we watch for the fingerprints: direct traffic from people who already know your name, branded searches climbing, and — the honest gold standard — new callers who say some version of "I asked an AI and it recommended you." We build that question into your intake so the signal actually gets captured instead of vanishing.
Now the vanity metrics we ignore. We do not celebrate raw "AI visibility scores" from tools that invent a number with no method behind it. We do not count how many times your name appears across a thousand irrelevant prompts nobody in Christiansburg would ever type. And we do not treat a mention in an answer to a national, generic question as meaningful — a Christiansburg cleaning company being named in "best cleaning tips" is noise; being named in "who cleans move-outs near Christiansburg" is signal.
The whole point is a market this size rewards being the specific, well-described local answer. A metro has fifty plausible names an AI could surface; the New River Valley has a handful, and if your inputs are clean and your reviews are real, you are one of them. We measure toward that reality, not toward a dashboard that flatters. See the fuller approach on our AI Search page.