Capabilities
Everything ClinicRadar does
Listed in the order a clinic meets them. Each capability produces evidence the next one can use.
The assistant
Do anything the interface does by asking — with every change shown to you for approval before it happens.
- Sets a clinic up from its website: name, address, phone, treatments, filled in and proposed.
- Reads your own data to answer, and says a score is withheld rather than estimating one.
- Every change is proposed with the exact values it would write, and nothing happens until you approve it.
- It cannot delete anything, and it can do nothing your own role does not already allow.
- Attach a photo or a PDF, and ask in English or Turkish.
Reputation intelligence
A deterministic Trust Health Score, a separate Coverage Confidence, and a themed breakdown of what the public actually says.
- Versioned, reproducible score — the formula version is shown next to the number.
- Coverage Confidence displayed separately; insufficient coverage suppresses the score.
- Themes derived from a reviewed clinic taxonomy in English, Turkish, German, and Arabic.
- Every theme opens onto the public items that produced it.
Crisis and risk detection
Four risk states, alerts that name the reason, and an incident flow that survives a handover.
- Risk states are exactly Calm, Watch, Alert, and Crisis — always with a label and an icon, never colour alone.
- Alerts carry the signal that triggered them and the evidence behind it.
- Incidents move New → Acknowledged → Investigating → Actioned → Resolved, with an audit trail.
- Notification recipients and thresholds are configured per clinic.
Competitor intelligence
Comparison against the competitors you name, on the same measurements, over the same window.
- You choose the comparison set; there is no opaque market average.
- Same score version and same coverage rules applied to every clinic in the set.
- Movement over time, not just a snapshot.
- Competitor evidence is public and linked, exactly as your own is.
AI answer-engine visibility
What answer engines say when a prospect asks about your treatment, your city, and your competitors.
- Prompts you define, run on a schedule within your plan's monthly allowance.
- Whether you appear, how you are described, and which domains are cited.
- Tracked across the languages your patient markets actually search in.
- Responses are stored as evidence with their capture date.
Action Center
A ranked, finite list of things to do, each attached to the finding that justifies it.
- Actions flow Suggested → Accepted → In Progress → Completed → Impact Review.
- Terminal states for Dismissed, Not Applicable, and Duplicate, so the list stays honest.
- Each action links to its evidence and to the measurement it is meant to move.
- Sized for a small team — the list is prioritized, not exhaustive.
Weekly and monthly reports
A weekly operating summary and a monthly executive report, delivered as PDF and shareable securely.
- Generated on a schedule, not on demand-only.
- Available in English, Turkish, German, and Arabic, with right-to-left layout support.
- Shared through authenticated routes or short-lived links — never a public URL.
- Includes the score version and coverage so a reader can judge the basis.