How we measure whether AI search is actually working for you
AI search is new enough that a lot of what gets reported about it is noise. So before we take on the work, we tell you exactly how we'll judge it — and which numbers we deliberately ignore, because chasing them wastes your money. The honest truth is that AI visibility is harder to measure than a Google ranking, but it is not unmeasurable, and pretending otherwise is how agencies hide behind fog.
The real question is straightforward: when someone asks an AI tool a question a Purcellville customer would actually ask, does your business come up, and is what it says accurate? So that is what we test. On a set schedule, we run the prompts your buyers really use — "who's a reliable [your trade] near Purcellville," "best [your service] in western Loudoun," "plan me a day in Loudoun wine country" if you're in tourism — across ChatGPT, Google's AI overviews, and Perplexity. We record three things every time: whether you're mentioned, whether the details are correct, and whether a competitor is recommended in your place.
From there we track movement over time. Are you appearing in more of those answer sets than you did last month? When you appear, are your hours, service area, and specialty described right, or is the AI repeating stale information from an old listing? Is the AI linking to your site as a source, or pulling from a directory that misrepresents you? These are the metrics that connect to whether a real person ends up on your page or on the phone.
Here is what we do not chase, because it looks impressive and means little:
- Raw "AI mention counts" with no context — being named in an answer nobody asks is not visibility.
- One-off screenshots of a single lucky result, which say nothing about whether you show up consistently.
- Generic "AI traffic" totals that can't be traced back to searches your actual customers make.
- Vanity comparisons against national brands you were never competing with for a Purcellville job.
What ties it all together is that AI answers are downstream of things you can control and we can watch: clear, well-structured pages, consistent business information, real reviews, and content that plainly answers the questions people ask. When those improve, AI visibility follows, and we can show the before-and-after in the actual answers — not a dashboard number you have to take on faith. That overlap is why our AI search work and our local SEO reinforce each other; the same clean foundation feeds both.
Because this field moves fast and the tools change month to month, we re-run the tests regularly and adjust. You get a plain-language report showing the prompts we tested, where you stood, and what moved — so you can see for yourself whether the work is landing, instead of trusting a metric that sounds good but proves nothing.