Getting started

Before you beginCopy link to section

You need three things: an email address you can receive mail at, the name and city of the clinic you want to monitor, and at least one public profile URL you know belongs to that clinic.

You do not need, and must never provide, any patient information. See Privacy and patient data.

The order things happenCopy link to section

1. Create an account. Sign up with email and password. You must verify your email before you can create a workspace or start a trial.
2. Create a workspace. A workspace represents your organisation. It holds your clinics, your team, and your subscription. See Your workspace.
3. Start a trial. Fourteen days, card required, one per organisation and per verified business domain.
4. Add a clinic and its locations.
5. Confirm your source profiles. This is the step that determines whether everything after it is correct. Do not rush it.
6. Choose patient-market languages and competitors.
7. Wait for the baseline. ClinicRadar collects your public history, classifies it, and produces the first score, coverage reading, and findings.

How long the baseline takesCopy link to section

It depends on how much public history exists and how many sources are involved. You are told when it starts and notified when it completes. Until it does, screens that depend on it say so rather than showing an empty state that looks like a result.

What to look at firstCopy link to section

Open the score, then immediately open the coverage reading next to it. If coverage is Insufficient, the number is deliberately suppressed — that is the product telling you there is not enough public evidence to say anything comparable yet, which is itself the finding.

Then read the themes, and open the evidence behind the two largest ones. That is usually enough to know whether the first action in the Action Center is the right one.

If something looks wrongCopy link to section

The overwhelmingly most common cause is a source profile attributed to the wrong business. Check Sources and profile confirmation before concluding the analysis is wrong.