How We Detect Healthcare Technology Across 87,000+ Clinic Websites

Every week, our crawler visits each clinic website and analyzes the publicly visible source code, network requests, and page structure. Nothing is hidden or proprietary about the inputs. We read the same HTML, JavaScript, and DNS records that any browser would load.

We look for technology fingerprints: patterns in the page source, embedded widgets, script includes, and third-party integrations that identify specific software. Each technology leaves traces when it integrates with a website, and those traces are consistent enough to detect reliably.

Each technology has its own detection signature, built from studying how that software integrates with clinic websites. We test each signature against known installs to validate accuracy before adding it to the crawler.

We maintain 200+ detection signatures across all categories, and add new ones as new tools enter the market. When a new PMS or booking platform launches in Australia or the UK, we build a signature for it.

Detection is binary: either we find a clear signal, or we mark it as "not detected." We never guess.

"Not detected" does not mean a clinic is not using software. It means we found no public signal on their website.

Some systems leave no public footprint. Server-side-only tools, desktop-installed software, and systems behind login walls are invisible to web crawling. A clinic could be running Best Practice or Medical Director and we would see nothing, because those systems do not embed any code on the public website.

This is why 48% of AU clinics show no detectable PMS, even though most clinics do use one. The absence of a signal is itself useful data: it tells you which clinics are not yet using a web-connected PMS, or are using one that does not integrate with their public site.

We crawl every week, not once. This means our data shows changes over time.

When a clinic switches PMS, adds a new booking system, or drops a payment tool, we detect it in the next crawl. The previous week's snapshot is compared automatically against the current one.

These migration signals are one of the most valuable parts of our dataset. They tell you which clinics are actively evaluating new technology. A clinic that just removed Cliniko and added Nookal is a very different sales prospect than one that has been on Cliniko for three years.

Weekly crawling also catches new clinics opening (new domains appearing), clinics closing (domains going offline), and gradual technology adoption trends across the market.

We only analyze publicly visible website data. We do not log into any system, access patient records, or scrape private information.

We do not use fake accounts, social engineering, or any form of unauthorized access. There are no bots filling out forms, no password attempts, no hidden browser sessions.

All data comes from what any visitor would see when loading a clinic's public website. If you can see it in your browser's developer tools, we can detect it. If you cannot, neither can we.