PatientNotes
PatientNotes is an AI scribe for solo clinicians and small practices, co founded in 2023 by a dentist, Dr. Alisa, and her brother Vadim Kouznetsov, who runs the company from an address in London. The same small team also builds other AI products, among them VetVisit AI for veterinarians and Dial Milo, a receptionist for small businesses. A clinician records the visit on the iPhone app, which keeps recording with the screen locked, or in a web browser, and moments after the visit ends PatientNotes produces a structured note in the clinician's template with suggested ICD-10 and CPT codes, patient instructions and follow ups. The clinician reads it, edits anything, and copies it into the EHR.
Templates come built in for 37 specialties, from chiropractic, dentistry, psychology and physical therapy to family medicine, pediatrics, cardiology and orthopedics, and clinicians can edit them or describe their own format. The company says more than 500 healthcare providers use it, and its pitch is aimed squarely at practices that most AI scribes, built for hospital systems, overlook.
Pricing is simple: the Team plan is $50 per user per month billed annually, or $45 for practices with five or more clinicians, with unlimited visits, a seven day free trial without a credit card and no contracts or setup fees; enterprise plans are quoted. Every paid plan includes a business associate agreement, patient data is stored in US data centers, isolated by customer and encrypted, recordings are deleted within a day, and the company says patient data is never used to train AI models.
Capability Axes
An AI Health Index grade measures what a buyer can verify from public sources on the date shown. It is not a rating of how good the product is. A vendor can build an excellent system and grade low on an axis because it publishes nothing an outsider can check. How grades read
The scribe is the product. PatientNotes listens to the visit through the iPhone app or the browser, understands clinical terminology without dictation commands, and turns the conversation into a structured note in the clinician's template, with suggested ICD-10 and CPT codes, patient instructions and follow ups drafted from the same visit.
The templates for 37 specialties, the phone app that records with the screen locked and the EHR export exist to make that output usable in a small practice's day. Without the speech and language models nothing remains, and PatientNotes sells itself as a replacement for human scribes and voice dictation rather than as a feature of other software.
Every note is a draft the clinician reads, changes as needed and copies into the EHR, so nothing reaches the record without the clinician placing it there. Suggested codes, patient instructions and follow ups arrive the same way, as suggestions alongside the note.
The review step carries all the weight, and nothing helps it along: no link from a sentence in the note to the moment in the visit it came from, and no flag on content the model was unsure about. Patient instructions are the output most likely to leave the practice unchanged, and their review is left to the clinician's habits.
Nothing identifies what produces the notes. PatientNotes says the scribe knows medicine and understands clinical terminology, works from natural conversation with no dictation commands, and fits notes to each clinician's edited templates or a format the clinician describes in words, and it names no speech recognition engine, language model, provider or version.
The iPhone app records even with the screen locked and syncs with the web app, which describes the capture side well. There are no release notes or change history, though, so an update to the model would go unannounced; outputs and workflow are described in detail, the underlying technology not at all.
The chain is described only in outline: patient data sits with infrastructure providers in US data centers, and AI models process it under a commitment that patient data is never used to train them. No cloud host, speech recognition service or language model provider is named, and there is no subprocessor list.
The business associate agreement included with paid plans is where downstream parties would be bound, and it is not posted. A practice therefore cannot see which outside companies hear its patients' visits before signing up.
No named practice and no measured result stand behind the product. The company says more than 500 healthcare providers use PatientNotes and that a clinician's first note takes under five minutes to set up, and the only testimonial on the site comes from its own co founder, a dentist.
Nothing reports time saved, note accuracy, code accuracy or how often clinicians rewrite drafts, and no customer in any of the 37 specialties is named. A practice comparing it with other scribes has the company's description of the product and the free trial to go on.
Patient data is never used to train AI models, PatientNotes states, and it is encrypted at rest with 256 bit AES and in transit with TLS 1.3, isolated from other customers' data and logged at every access.
Retention is set out by data type. Visit recordings are deleted within 24 hours, once transcription is done; transcripts and notes are kept for the length of the subscription plus 30 days unless a clinician asks for earlier deletion; account details are kept up to seven years after termination. De identified data may be kept indefinitely for service improvement, without a description of how it is de identified. Safety engineering is limited to the clinician's review of each draft, with no account of how invented findings or wrong codes are prevented.
A business associate agreement is included automatically with every paid subscription, Team and Enterprise, at no extra cost, and PatientNotes will sign a practice's own agreement. Its terms of service state that PatientNotes operates as a business associate and incorporate the agreement by reference for anyone using the service with protected health information, and HIPAA Security Rule safeguards, regular risk assessments and documented policies are described.
The agreement is not posted, and the two statements differ in reach: the promise of an agreement covers paid customers, while the terms reach every user, which leaves the free trial's position unclear. Whether the agreement binds the infrastructure and AI providers that handle patient data, and which providers those are, is not stated.
PatientNotes says outside firms carry out regular security audits and penetration tests, with no frequency, firm or summary of results given, and holds no attestation of its own. The controls it describes are 256 bit AES encryption at rest, TLS 1.3 in transit, audit logs on every access to patient information, data isolation between customers, workforce training, access management and regular risk assessments, and it points to the compliance certifications of its infrastructure providers, which belong to the hosts rather than to PatientNotes.
No SOC 2 report, HITRUST certification or ISO 27001 certificate is claimed, and there is no trust center or report to request. That outside testing is the one independent check, with its scope and dates unstated.
PatientNotes sells documentation and says nothing about regulatory status. Its sample note includes an assessment and a plan, with a medication and follow up, but those are drafted from what the clinician said during the visit and reviewed before use.
Patient instructions are also drafted for the clinician to check and hand over. The terms describe the service as generating clinical documentation such as SOAP and progress notes and make the clinician responsible for its accuracy and completeness, which is the nearest thing to a scope statement. Nothing markets PatientNotes as diagnosing or recommending treatment, so the documentation positioning matches the product, though no intended purpose statement sets that boundary in writing.
Nothing describes how the scribe is evaluated or governed. PatientNotes says it changes based on what clinicians ask for, which describes how features are chosen rather than how model behavior is tested, and its comparison pages against Freed, Heidi, DeepScribe, Suki, Nuance DAX, Dragon and Abridge set out features rather than measured quality.
The product spans 37 specialties, from dentistry with its tooth by tooth findings to psychology and pediatrics, and no results are offered for any of them, or by accent or language. A practice cannot tell whether notes in its specialty are as reliable as the sample shown.
The clinician's review is the check, and access to patient information is logged so a practice can see who opened a note. Support goes directly to the team that builds the product rather than a call center, which shortens the path when something goes wrong and lets clinicians ask for template fixes.
PatientNotes offers no error rate, warranty or remedy for the notes, codes or patient instructions it drafts. With no contract and cancellation at any time, a practice's main recourse for poor output is to stop paying, and responsibility for every signed note stays with the clinician who copied it into the record.
No record system integration is offered. Clinicians copy the finished note into their EHR, and the Team plan lists EHR export without naming a system, file format or interface; the enterprise plan adds an admin dashboard but no integration is described for it either.
For the solo and small practices PatientNotes targets, copy and paste works with any system, dental, chiropractic or medical, and setup takes minutes rather than an IT project. It also means nothing flows from the chart into the note, such as the patient's history, medication list or prior visit, and nothing is filed back to the record automatically.
All patient data is stored in data centers in the United States, and each customer's data is isolated from every other organization's. The service is a single hosted product, used through the iPhone app or a browser, with encryption at rest and in transit.
What is left open is the processing path: the cloud provider is not named, and nothing says where recordings are transcribed or which AI providers process them, so a practice cannot confirm that every step stays in the United States. PatientNotes gives its company address in London and serves clinicians in the United Kingdom and Australia as well, which makes where staff and processors reach patient data a fair question for a US practice.
Prices are listed and the product can be bought without a sales call. The Team plan is $50 per user per month billed annually, $600 a year, or $45 per user for practices with five or more clinicians, with monthly billing available at a higher rate; annual billing saves 29 percent. It includes unlimited visits, custom templates, all note formats, priority support, EHR export and a business associate agreement.
A seven day free trial needs no credit card, and there are no contracts or setup fees, with cancellation at any time. Enterprise plans for larger practices and health systems, with unlimited users and an admin dashboard, are quoted.
Small outpatient practices are the stated market, across 37 specialties with built in templates: chiropractic, dentistry, psychology, podiatry, occupational and physical therapy, family and internal medicine, pediatrics, obstetrics, urgent care, orthopedics, psychiatry, dermatology, cardiology, neurology, ENT and ophthalmology among them. Each template matches a visit type, such as a periodontal exam or a well child visit.
No practice in any of those specialties is named as using the product, so the breadth rests on the template library rather than on deployments. Hospital and inpatient use are explicitly not the focus.
Alternatives to PatientNotes
Vendors in the same categories as PatientNotes whose graded profiles look most alike, with a line on what each product does.
Similar is not the same, so check that each one does the job you need. No vendor pays for placement.
Noterro, from Noterro Inc. in Canada, is practice management software for allied health practitioners, built for massage therapy, chiropractic, physiotherapy, osteopathy, mental health and…
Medical Scribe, sold at medicalscribe.app by Leveled Platforms, Inc., is an AI scribe that records a patient visit or a clinician's dictation and turns it into a note ready for the record.
Cerbo AI Scribe is the ambient documentation feature built into Cerbo, an EHR and practice management system for functional and integrative medicine, direct primary care, cash based…
Medscape Scribe is a free ambient documentation tool launched in October 2024 by Medscape, the clinical information service, available inside the Medscape mobile app and website that a very…
MedWriter is an AI medical scribe from S7Labs, LLC, doing business as Medwriter, built for outpatient clinicians and strongest in psychiatry, primary care, neurology and pain medicine.
Pabau is a practice management platform for private clinics and medical spas, and the artificial intelligence product inside it is Pabau Scribe, an ambient medical scribe that generates…
Ordered by how closely the graded profiles match across the fifteen axes, inside the same categories. See how each axis is graded in the methodology.
Pricing
Vendor-published figures are labeled as such. Figures labeled “Estimated” are derived from third-party sources and have not been confirmed by the vendor.
| Entry Price | Pricing Basis | BAA Tier | Implementation | Source |
|---|---|---|---|---|
|
$50 per user per month (billed annually)
$50 baseline
|
Per user subscription, unlimited visits | Business associate agreement included with every paid plan (Team and Enterprise); not the free trial | None; no setup fees or contracts | Vendor Published |
The Team plan is $50 per user per month billed annually ($600 a year), or $45 per user for five or more clinicians; monthly billing costs more, with annual saving 29 percent. It includes unlimited visits, custom templates, all note formats, priority support, EHR export and a business associate agreement. A seven day free trial needs no credit card; cancel anytime. Enterprise plans with unlimited users and an admin dashboard are quoted.