Visibility tooling produces hundreds of numbers, most of which move on their own and mean nothing. This lesson picks five that actually answer the question you care about: are more of the right patients finding us.
Why fewer is better
A dashboard with forty metrics cannot be acted upon. Worse, it lets you find a rising line whenever you need one, which is how practices convince themselves something worked.
Pick a small set, record it monthly, and hold it stable so the comparisons mean something.
One: new patient enquiries, by source
The number that matters most, and the only one that is unambiguously a business outcome.
Count enquiries, meaning calls, forms, and booking requests from people who are not existing patients, broken down by the source field from chapter eight. Everything else in this list is a leading indicator of this.
If your enquiries are flat while your impressions have doubled, you have improved your visibility to the wrong people, or your listing is being seen and not acted on.
Two: profile actions
From your Business Profile, per office: calls, direction requests, website clicks, and booking clicks.
These are the closest thing to intent you can measure on a listing. Direction requests are particularly useful for a clinic, since nobody asks for directions casually. Track the total and the trend per office, not per day.
Three: directory coverage and accuracy
A count, not a rate: how many of your priority listings are present and correct.
Take the inventory from chapter three and score each entry as correct, wrong, or missing. Track the number correct over time. This is the only metric here that measures the work you are actually doing rather than its downstream effects, and it moves immediately when you fix something.
Include your payer directories, since those are the most frequently wrong.
Four: AI presence rate
From chapter five: across your fixed prompt set and the engines you track, in what share of runs are you named.
Record it as a rate across repeated runs rather than a yes or no, and note whether the details given were correct. A practice named half the time with a wrong phone number has a more urgent problem than one never named at all.
Five: reviews, count and recency
Two figures: total reviews and how many arrived in the last ninety days.
The second matters more. Recency feeds both local prominence and the sentiment summaries assistants produce, and it is the metric that reveals a review habit has quietly died. A practice with two hundred reviews and none in eight months is declining, and the total alone hides that.
What to leave off
- Rank position as a headline number. It varies street by street, so a single figure is meaningless. Chapter nine has a lesson on reading grids properly.
- Total website traffic. Too easily inflated by irrelevant visitors.
- Impressions and views alone. A leading indicator at best, and a vanity metric at worst.
- Social followers. Not a driver of local visibility.
- Keyword counts and citation counts. Volume metrics that reward the wrong behaviour.
- Bounce rate, which is routinely misinterpreted.
How to record them
- Monthly, on a fixed date, in one place.
- Per office, not just practice-wide, since offices diverge.
- Alongside a change log noting what you did and when, which is what makes attribution possible at all.
- Capture a baseline before making changes. Without it, no later claim of improvement can be substantiated.
The following lessons cover where each of these numbers comes from, starting with the profile reports.