Editorial · Methodology v2026.07

Methodology.

Every scribe is scored on six dimensions with fixed weights, published before scoring. Testing uses four real visit types across independent primary-care and specialty practices with 1–50 clinicians.

§ 01 / The rubric

The six dimensions.

HIPAA & security

20%

Signed BAA, data encryption at rest and in transit, audit logging, US data residency, retention controls, SOC 2 or HITRUST posture.

Note quality

22%

SOAP fidelity across primary care, urgent care, and common specialties; ICD-10 and CPT coding assistance; hallucination rate on real transcripts.

EHR integration

18%

Native or SMART on FHIR integration with Epic, Athena, eClinicalWorks, DrChrono, Elation, Practice Fusion, Kareo, NextGen, and other small-clinic EHRs.

Workflow fit

15%

Setup time, training burden, template customization, review-and-sign flow, mobile capture, offline behavior, multi-clinician handoff.

Pricing & contract

15%

Per-clinician cost, group discounts, trial length, month-to-month vs annual, transparent tiers, hidden usage fees.

Support & reliability

10%

Documented uptime, human support hours, onboarding assistance, response time for clinical bugs, status page transparency.

§ 02 / Testing protocol

Four visit types, hand-scored.

Each scribe is tested on four transcript sets: an acute primary-care visit, a chronic-care follow-up, a well visit, and a common specialty visit (rotated quarterly between psychiatry, women's health, dermatology, and orthopedics). Transcripts are hand-scored against the true visit content for coverage, hallucination, and structure.

§ 03 / Refresh cadence

Verified quarterly.

The full ranking is re-verified at least once a quarter. Individual scribe pages carry the date they were last verified. A change of more than 0.3 in overall score triggers a re-test.

Read the independence statement for our commercial policy.