AI Visibility
How to Measure Whether Your AI Visibility Is Actually Improving
Once a practice starts fixing its five AI visibility signals, the next question comes fast: is any of this working? It's a fair question, and a hard one — because the old ways of measuring marketing progress, rankings and website traffic, have quietly stopped telling the truth.
Here's why. As of 2026, SparkToro's analysis found that fewer than one in three Google searches still ends in a click to the open web — roughly 68% of searches are now 'zero-click', and for searches that trigger an AI Overview the zero-click rate climbs toward 83%. Your website traffic can fall while your visibility rises, because the patient got their answer — with your name in it — without ever clicking through.
Why traffic and rankings mislead you now
Pew Research put concrete numbers on the shift: when an AI summary appears on a Google result, users click a traditional link in just 8% of visits, versus 15% without one. So a practice can be mentioned more often in AI answers, win more consideration, and still see flat or falling click counts in its analytics. If traffic is your only scoreboard, you'll conclude the work failed at exactly the moment it started paying off.
The fix is to stop measuring only the click and start measuring the mention. That means adding a few signals your analytics dashboard was never built to capture — and checking them on a regular cadence.
The direct test: ask the assistants yourself
The single most useful measurement is also the simplest, and it costs nothing. Once a month, open the tools your patients actually use and ask the questions they actually ask, then write down what comes back.
- In Google, search your core queries — 'family dentist in [your city],' 'emergency dentist near [neighborhood],' 'dentist that takes [common local insurance]' — and note whether you appear in the AI Overview, not just the map pack or the blue links.
- In ChatGPT, ask the same questions and record whether your practice is named, and how it's described. With roughly 900 million people using ChatGPT weekly, this is no longer a fringe channel.
- In Perplexity, ask the research-phase questions — cost, procedures, second opinions — and see which practices get cited.
- Note not just whether you appear, but the words the AI uses about you. Those words tell you which of your signals are being read.
Keep the results in a simple spreadsheet: the query, the assistant, whether you appeared, and what it said. Do it the same way each month. Within two or three months you have a trend line no analytics tool can give you — a direct read on whether AI is recommending you more often.
The leading indicators worth tracking
Alongside the direct test, track the upstream signals that AI reads — because these move before your recommendation rate does. The most useful ones are the same fundamentals that drive visibility in the first place: your Google Business Profile insights (calls, direction requests, and profile views, which capture the patients who never touch your website) and your review cadence (new reviews per month and your response rate).
- Review velocity: new reviews per month and the trailing three-month average. A rising line here reliably precedes rising AI visibility.
- Google Business Profile actions: calls, direction requests, and website clicks from the profile — the patient activity that happens entirely outside your website analytics.
- Citation consistency: a quarterly check that your name, address, phone, and hours still match across Google, Yelp, Healthgrades, and your own site.
- New-patient source: the single most honest metric of all — ask every new patient how they found you, and watch the share who say 'I searched and your name came up.'
Ask new patients how they found you
The most valuable measurement tool in your practice is the intake form and a single question at the front desk: how did you hear about us? It's low-tech and slightly imperfect — patients don't always remember precisely — but tracked over time it cuts through every measurement problem AI has created. When the share of new patients who found you through search and AI answers climbs quarter over quarter, that is the result. Everything else is a leading indicator for that one number.
Put together, these measures replace the single misleading scoreboard of website traffic with a small dashboard that actually reflects how patients find a practice in 2026. If setting that up from scratch feels like a lot, it's exactly what the AI Visibility Center is designed to short-cut — a baseline read today, so next quarter's numbers have something to be measured against.
Frequently asked questions
- Why is my website traffic flat if my AI visibility is improving?
- Because most searches no longer end in a click. As of 2026, roughly two-thirds of Google searches are zero-click, and for AI Overview results it's higher still. Patients increasingly get your name and details inside the answer itself, so you can gain visibility and consideration without a corresponding rise in website visits. That's why the mention — not the click — is what you have to measure.
- How often should I check my AI visibility?
- Run the direct test — asking the assistants your core patient questions — once a month, and do a fuller review of your leading indicators and citation consistency once a quarter. The signals don't change overnight, but a monthly rhythm catches drift early and builds the trend line that makes progress visible.
- Do I need special software to measure this?
- No. The most reliable measurements are a monthly spreadsheet of what the assistants say when you ask, your existing Google Business Profile insights, your review counts, and a 'how did you hear about us?' question at intake. Paid AI-visibility tracking tools exist and can help at scale, but the fundamentals cost nothing but a little consistent attention.
- What's a realistic timeline to see measurable improvement?
- Consistency and structure fixes can show up in AI answers within a few weeks. Review velocity and authority compound over six to twelve months. If you baseline now and measure monthly, you'll usually see the direct test — how often you're named — start to move within a quarter, well before it shows up in new-patient numbers.
Take the next step
Get a baseline before you measure the progress.
A free AI Visibility Score gives you a dated snapshot of how AI sees your practice today across all five signals — the starting line every future measurement is compared against, plus a 30-day plan to move it.
Get my AI Visibility Score