Skip to main content
NewNMC’s advertising guidelines for doctors are in force from 6 Oct 2026. See what to change, and get a free compliance auditNew NMC advertising rules for doctors: what to change
Website

Doctor Profile Pages at Scale: Schema, Freshness and Verification for 500 Consultants

A field model, schema pattern and update workflow for hospital doctor profiles, so registration numbers, qualifications and availability stay accurate across hundreds of consultants.

Doctor Profile Pages at Scale: Schema, Freshness and Verification for 500 Consultants

A hospital with five hundred consultants does not have a content problem on its doctor pages. It has a data problem. A profile that shows a surgeon who left in March, a qualification typed three different ways, or "available today" for a clinic that was cancelled is worse than no profile, because the patient acts on it.

This post is for hospital web teams, digital heads and CMOs who own or are about to build a doctor directory. It sets out a field model, the schema.org markup that should sit on top of it, and an update workflow that keeps registration number, qualifications and availability accurate without relying on someone remembering to edit pages. It is a checklist-style build guide, not a design brief.

One boundary first. Every clinical and credential statement on a profile must come from the doctor and be confirmed by your credentialing or medical-affairs team. Marketing owns the template and the workflow; it does not own the facts. Your NMC-related wording rules are covered in the NMC compliance handbook.

Treat the profile as a record, not a page

The first design decision is where the truth lives. If each profile is a free-text page in a CMS, you have 500 places for errors. If each profile is rendered from a structured record, you have one place per field and one validation step.

A workable record has four groups of fields.

GroupFieldsOwnerPublic?
Identityfull name as registered, display name, photo, gender (optional), languagesDoctor, HRYes
Credentialsregistration number, registering council, qualifications with institution and year, designationCredentialing officeYes, after verification
Practicespecialty, sub-specialty interests, conditions and procedures treated, department, hospital and locationDepartment headYes
AvailabilityOPD days and times by location, appointment link, accepting new patients, last-confirmed timestampScheduling team or HIS feedYes, with timestamp

Keep two further groups that are not public but drive the workflow: a verification block (who checked, when, against what) and an editorial block (status, last clinical review date, next review due).

Use structured values wherever possible. Qualifications should be a list of objects (degree code, institution, year), not a string like "MBBS MD (Med) DM Card AIIMS". Registration should be two fields, number and council, never one. Specialty should come from a controlled list so that "Cardiology", "Cardiologist" and "Heart Specialist" cannot coexist as three different filters.

Registration number: display it, verify it, date it

Showing the registration number helps patients and referrers check a doctor, and it is the field most likely to be wrong, because it is copied by hand from certificates.

A sensible rule set:

  • Store the number and the council exactly as they appear on the registration certificate. Do not normalise or reformat on the way in.
  • Verify each new number against a public register before the profile goes live. The National Medical Register is the NMC-maintained register of registered doctors, and state medical council registers remain the other check. Check what each register currently shows and record the check.
  • Where the doctor has an ABDM Healthcare Professionals Registry ID, the Healthcare Professionals Registry is another place to cross-check. Treat it as optional metadata, not a replacement for the council number.
  • Record the verification as data: verified by, verified on, source checked. A profile with no verification date should not publish.
  • Re-verify on a schedule and on events, such as a change of name or a new qualification, not only at onboarding.

Do not publish a number you could not verify "temporarily". The temporary state is the one that becomes permanent.

Schema: Physician, with the right restraint

For individual doctors, the natural type is Physician. It sits under MedicalOrganization and LocalBusiness, which is why it can carry address, telephone and opening hours as well as medical properties. The Physician type itself adds availableService, hospitalAffiliation and medicalSpecialty, and isAcceptingNewPatients comes from MedicalOrganization. The usNPI property is US-specific; leave it out.

A practical mapping from the record to markup:

Record fieldschema.org propertyNote
Display namenameMatch the visible heading
PhotoimageSame image as on the page
SpecialtymedicalSpecialtyUse the schema.org enumeration where a value exists
HospitalhospitalAffiliationPoint to the Hospital entity defined once on the site
Procedures treatedavailableServiceOnly those the doctor performs
OPD hoursopeningHoursSpecificationPer location; omit if not maintained
Accepting patientsisAcceptingNewPatientsOnly if the source is current
External profilessameAsCouncil register page, if it has a stable public URL
Registrationidentifier as a PropertyValuename: council, value: number

Three rules keep this honest. First, Google's structured data policies say not to mark up content that is not visible to readers, and to provide up-to-date information, so every property above should correspond to something a visitor can see. Second, define the Hospital and the departments once, with stable @id values, and reference them from every doctor; do not re-describe the hospital 500 times. Third, there is no rich result promised for a Physician type in itself, so the value is clarity for search and AI systems, not a badge. The wider approach to medical markup is in our advanced medical schema guide and the schema markup service page.

Do not add review or aggregateRating markup copied from third-party platforms. If you display your own patient feedback, handle it under its own rules and keep it separate from the credential block.

Want this applied to your practice? Get a free website audit.

Book my free audit

Freshness: design for decay

Profile data decays at predictable moments. Plan around those moments instead of a vague "review yearly".

Trigger-based updates. Connect the profile record to the events that change it:

  • Joining and exit: HR or credentialing raises the event; the exit event unpublishes the profile and sets a redirect to the department page the same day.
  • New qualification or designation: credentialing updates the structured field and re-verifies.
  • Location or roster change: the scheduling team updates OPD fields, or the HIS feed does.
  • Leave: a temporary banner field with an end date, which expires itself.

Scheduled reviews. Give each field group its own interval. Availability may need a feed or a weekly check; credentials a yearly re-verification; the free-text biography a review when the doctor or department asks. Store next_review_due per profile and generate a weekly list of overdue records for the owner.

Availability is the dangerous field. If you cannot feed it from the scheduling system, do not display exact slots. Show OPD days in general terms with a "last confirmed" date, and send the patient to the appointment step for live slots. If you can feed it, show the source time. The booking side is covered in appointment scheduling.

A review workflow that scales to 500

Four states are enough: draft, in clinical review, live, and stale. A profile moves from draft to review when the doctor submits it, from review to live when credentialing and the doctor both approve, and from live to stale automatically when next_review_due passes or a trigger event fires. A stale profile stays visible only if the stale field is non-critical; a stale registration or exit flag hides it.

Assign a department-level owner, not a central marketing owner, for the content the department understands. Marketing runs the queue. A department coordinator who handles 25 doctors will keep it current more reliably than one person trying to chase 500.

Build three automated checks into the pipeline:

  1. Schema validation on every deploy, with a test that fails the build if a live profile lacks name, specialty, hospital or a verified registration.
  2. Visible-versus-markup check: the rendered page and the JSON-LD must agree on name, specialty and hospital.
  3. Link and duplicate check: one profile per registration number, and no two profiles pointing to the same photo or biography.

A single-source record also makes a multi-location website easier, because a doctor who sees patients at two sites is one record with two practice entries, not two pages that drift apart.

Hypothetical example of the field model in use

Take a 400-bed hospital in Pune with a cardiology department of 18 consultants. The department coordinator is the named owner. Each consultant has one record; two of them practise at a satellite centre and carry two practice entries. When a consultant resigns, HR's exit event unpublishes the profile and redirects it to the cardiology page. When another passes a fellowship, credentialing adds the qualification and re-verifies, and the page updates without a web team ticket. The numbers here are illustrative; the point is that no step depends on someone remembering to open a page and edit it.

Pre-launch checklist

  • Every live profile has a verified registration number, council and verification date.
  • Qualifications are structured data, not a text string.
  • Specialties come from a controlled list.
  • Physician JSON-LD matches visible content, and the Hospital is defined once.
  • Availability is either fed from a source with a timestamp or shown only as general OPD days with a "last confirmed" date.
  • An exit event unpublishes and redirects within a defined time.
  • Each profile has an owner, a status, a last clinical review date and a next review due date.
  • Deploys fail when a live profile is missing required fields.
  • All clinical and credential wording has been approved by the provider, and advertising wording follows NMC rules.

If you are planning a directory or rebuilding one, our provider bio pages service covers the template and workflow side.

Filed underdoctor profile page schemahospital doctor directory websitephysician schema markupdoctor profile page templateconsultant profile pages hospital

About the author

Co-Founder & CTO · Gurugram, India

Arush builds the systems the rest of the company runs on — the patient-acquisition pipelines, the AI chatbots, the analytics that tie a single click to a booked appointment. If a campaign reports a number, it's because something his team built is tracking it.

Free guide · 13 pages

WhatsApp AI Chatbot for Hospitals (14-Day Deploy)

The 14-day WhatsApp booking-bot implementation playbook.

The Pioneers Brief — one essay, three benchmarks, one teardown, every Thursday. Subscribe free →

Free website audit

Your next chapter starts with a conversation.

Three quick questions. We come prepared with a plan for your specialty and city.

  • A 30-minute call with a healthcare specialist
  • Your top three growth opportunities, in writing
  • Honest advice — even if that means not hiring us
  • No obligation and no hard sell
You’ll speak with a senior strategist
Consultations are hosted by Arush Thapar’s team.
Step 1 of 3Takes 30 seconds

What do you want to improve first?

Details stay privateNo obligationReply within 1 working day

Client portfolio · Healthcare websites

Work relevant to doctor profile pages at scale schema freshness and verification.

Explore client work →