Attio for healthcare

• A person is not a patient. Identity, marketing email history and form submissions belong on the Person record; active care belongs on a separate Patient record. Merge them and a lead who never converted shows up in retention reports, while a patient who finished care still gets newsletters about being a lead.
• Each visit is its own atomic record. A visit has a date, a type, a price, products purchased and notes. Stuff those into fields on a contact and the third visit overwrites the second, so revenue per patient becomes a guess.
• Recurring follow-ups depend on date triggers, not memory. A patient due next quarter does not surface unless the system schedules it, and CRMs that rely on someone remembering to set a reminder lose a meaningful share of retention every cycle.
• B2C and B2B run on different rhythms inside the same business. Individual leads convert in days; corporate partnerships with insurers, banks, employers and clubs move over months with proposals and committee approvals. Force both through one pipeline and one of the two teams is always working in the wrong shape.

• Custom objects as first-class citizens. Patient and Examination have to be their own objects with their own workflows, not awkward extensions of Contact and Deal. Where custom objects are a paid add-on or a database afterthought, the model breaks within three months.
• Many-to-many relationships. A patient buys multiple products and a product is bought by multiple patients; a B2B company has multiple key contacts; a doctor sees patients across multiple insurance partners. Forcing these through one-to-many workarounds creates phantom records and broken reports.
• Scheduled workflows, not only trigger-based ones. Retention needs "every Monday, look at patients due in 30 days and move them to the right list" — a different primitive from "when someone fills a form, send an email". A CRM that only does the second cannot do recurring care.

That is the full model, and it runs without any healthcare-specific software.

• People — the source of identity and the journey from lead to retention. Status with four stages (Lead, Warm Lead, Ongoing Patient, Retention), a source field tagged by acquisition channel, a type field for B2C versus B2B contact, and email and phone history anchored on the record, deduplicated on email.
• Patients — a custom object created automatically when a Person flips to Ongoing Patient. Care-specific fields only: linked Person, next examination date, total lifetime cost rolled up from Examinations, number of past examinations, current treatment notes. Separate from People because People care about acquisition and Patients care about cadence.
• Examinations — one atomic record per visit: date, type, value for that visit only, products purchased, summary notes. The Patient record never overwrites; the history is the list of linked Examinations.
• Companies — B2B partners on their own pipeline: insurers, banks, employers, corporate health programmes. Company type, key contact, qualifying stage, deal value, package, channel and reseller. A corporate contact still gets a Person record, linked to the Company.
• Products — services purchased by patients, many-to-many with People for accurate revenue tracking, each with a price that rolls up into the Patient lifetime cost.

The trap most clinics fall into is picking one CRM for individual patients and a second for corporate sales: two tools, two sets of contacts, two subscriptions and no shared revenue picture.

The same People object handles both. A B2C lead moves through the patient journey on Status. A B2B contact carries a Type flag and a relationship to the Company. The kanban filters by Type, so the B2C team sees only individuals and the partnerships team sees only corporate.

Companies get their own pipeline with stages that match a longer cycle — Need, Choose, Worry, Commit — and a B2B dashboard tracking deal value by stage, channel and reseller. None of it leaks into the patient journey reports, and both pipelines feed the same revenue dashboard at month end.

The retention automation is the one that pays for the whole setup. A clinic that captures 70 percent of follow-up visits instead of 50 recovers the implementation cost in the first quarter on a base of 200 active patients.

• Form submission to Person, with no duplicates. If the Person exists the source field updates; if not, a new Person is created. A Slack approval gate on outbound email prevents anything going out unreviewed.
• Person becomes Patient automatically. When status flips to Ongoing Patient, a linked Patient record is created with defaults inherited from the Person.
• Lifetime value rolls up from Examinations. Each new Examination sums the cost and updates Total Value on the linked Patient, so revenue per patient stays accurate without manual entry.
• Examination type routes the patient into the right list — full health check or specific test — with no manual list management.
• Date-triggered retention staging. Scheduled workflows scan the retention list daily and move patients due in one month or one week into the right stage. Follow-ups never depend on memory.
• The post-visit form fills patient fields. On submission the workflow finds the matching Patient and populates the answered fields, with no re-keying.

Attio offers HIPAA compliance on the Enterprise tier with a signed BAA. That is real, and it is priced for funded teams rather than solo clinics.

The practical answer for most healthcare businesses we work with is to keep PHI out of Attio and use it for the journey and relationship layer. Names, emails, phone numbers, lead source, examination dates, examination type and revenue are not PHI on their own. Diagnostic data, treatment plans and clinical notes are, and those stay in the EHR.

That split works for preventive care, wellness clinics, health-tech with a B2C funnel and concierge practices. It does not work for hospital systems running clinical workflows in the CRM. If the line feels blurry for your operation, have the fifteen-minute conversation with your compliance officer before the implementation rather than after.

A core build — five objects, lead intake automation, retention staging, B2B pipeline and both dashboards — takes about nine working days. Longer when historical imports from Sheets, Brevo, Calendly and Gmail are in scope. The blocker is rarely Attio; it is the team agreeing what each pipeline stage means and what triggers a status change. Once that is settled, the build is mechanical.

• Hospital systems with clinical workflows in the CRM. Salesforce Health Cloud or Epic are the standards for a reason: integration with billing, claims and clinical data is the actual product.
• Practices that need a built-in EHR. Pabau, NexHealth and Practice Better bundle scheduling, charting and a patient portal. Attio is a pure CRM, and if the team will not run two systems, the bundle wins.
• Operations billing insurance directly through the CRM. Attio has no claims module; a third-party tool covers it, but the integration is real work.
• Single-doctor practices under 100 patients. A spreadsheet is fine until you cross that threshold, and setting up a CRM before the volume justifies it is cost without return.