Heidi vs MedMETs: three weeks of side-by-side use, honestly

Heidi has dominated Australian GP awareness of ambient AI scribes for the past 18 months. It's polished, it's fast, and it does the core job of generating a SOAP note from a consult well. I used it daily in clinic for three weeks before switching to MedMETs for the next three. Here's how the two compared.

Where Heidi shines

Heidi's note quality at the basic SOAP level is excellent. The transcription is fast, the structure is clean, and the integration with most Australian clinical software (Best Practice, MedicalDirector) is mature. For straightforward acute consults — "I've had this cough for four days" — Heidi produces a sign-off-ready note in seconds.

It's also brand-trusted: Heidi has been in the Australian market longer than any of its competitors and most of my colleagues have heard of it.

Where I felt the limits

Heidi is a scribe. That's the whole product. The note exists. What comes after the note is somebody else's problem.

For a complex chronic-disease patient, the scribe-only workflow meant I was generating the note, then opening a separate tool to draft the GPMP, then opening another to flag the 732 review, then opening the patient's app (which they don't have) to share follow-up instructions. The scribe saved me 6 minutes; the rest of the workflow ate them back.

Where MedMETs was better

Three concrete things. First, the note generation triggered the care-plan draft automatically when the patient's chronic conditions were referenced in the consult — so the GPMP was 80% drafted before I had to think about it. Second, the patient app meant I could share the plan and follow-up tasks before the patient hit the carpark. Third, the screening prompts surfaced overdue assessments during the consult — so a Heart Health Check that I'd have rebooked for next month got done in the same visit.

Where Heidi was better

Brand familiarity. My patients hadn't heard of MedMETs. Some of them asked questions about it that I had to answer. With Heidi, the AI scribe was invisible to them — which, depending on your stance on transparency, is either a feature or a problem.

Pure note polish on basic acute consults. Heidi's notes were marginally more polished on simple consults — the differential phrasing, the formatting consistency. The gap was small but it was there.

The honest verdict

Heidi is the right answer if you want a really good scribe and nothing else. MedMETs is the right answer if you want the scribe to be the entry point to the rest of the chronic-disease workflow — care plans, screening, billing, patient continuity. If you measure value as 'minutes saved on documentation', Heidi competes. If you measure value as 'practice-wide outcomes and revenue', MedMETs pulled ahead in week two.