Skip to main content

Doctor and Clinic Schema: Describe the Right Entity Without SEO Promises

Use structured data to describe the same doctor, clinic and page that patients can see. Keep the person separate from each physical practice location, choose types for their actual meaning and maintain stable identifiers. Google feature eligibility is a separate question: adding Person, MedicalClinic or ProfilePage markup does not guarantee rankings, rich results or AI citations.

By Bractis · Updated · 9 min read

Indian clinic owners, doctors and website teams reviewing profile markup without treating a schema validator as proof of search success.

Decide what the page is actually about

A doctor, a clinic and a page describing them are not interchangeable. One physician may consult at two premises, while one polyclinic may host several doctors. If the website gives every page the same identity, a change to one location's address can appear to move the doctor and the whole organisation together.

Start with a small entity map. Identify the professional, each relevant facility and the public pages that describe them. Decide which facts belong to each: qualifications and professional biography belong to the person; address and physical opening arrangements belong to a location; the page has its own URL and modification history.

Schema.org defines Person and MedicalClinic as different vocabulary types. The retrieved vocabulary pages identify themselves as development versions, so a developer should validate the chosen properties against the implementation's supported vocabulary. The important clinic decision is the identity distinction, not collecting the largest possible number of schema labels.

Do not add a schema type merely because a competitor uses it. A diagnostic service, an organisation and an individual clinician can have different meanings even when they share a name or reception desk.

Source: Schema.org: Person

Source: Schema.org: MedicalClinic

Separate vocabulary validity from Google feature eligibility

A JSON-LD block can be syntactically valid while describing the wrong thing. It can also use a legitimate schema type without meeting a particular Google search-feature policy. Review syntax, factual accuracy and feature eligibility as three separate checks.

Google's ProfilePage guidance focuses on a single affiliated person or organisation and gives examples such as employee and author pages. Its stated purpose relates to creators and their perspectives. A third-party listing or review page does not become eligible simply by being named a doctor profile. Assess the actual page and relationship before selecting that feature.

Google's LocalBusiness guidance covers business information such as locations and hours. That does not turn a clinic's general opening hours into a guarantee that a particular consultant is available, or establish that adding markup will enable direct booking from Search.

Ask the developer which supported feature they are targeting and why the page qualifies. A report saying only “all schema valid” leaves the most important editorial questions unanswered.

On a narrow screen, scroll the table horizontally to read all columns.

Doctor website entities and the facts they can describe
Entity or layerAppropriate factsCommon mistake
DoctorName, supported credentials and professional roleUsing the clinic's address as the doctor's only identity
Clinic locationReal premises, contact route and opening informationCombining two branches into one address
Profile pageThe page's focus and its relationship to the personAssuming every directory listing qualifies for a feature
ServiceThe service actually available at the stated locationClaiming all procedures at every branch
Booking stateCurrent service, slot and confirmation conditionsTreating static markup as a live reservation
Search appearanceObserved indexing and feature resultsCalling validation a ranking guarantee

Source: Google Search Central: ProfilePage structured data

Source: Google Search Central: LocalBusiness structured data

Match every public assertion to visible evidence

Build a claim sheet from the page before generating markup. For every name, qualification, language, address, fee and schedule, record who supplied it and when it was checked. A hidden field should not make a stronger claim than the visible biography.

Suppose the biography says a doctor offers general medicine consultations in Hindi and English at a Tuesday evening clinic. The markup should not invent cardiology expertise, Punjabi fluency or seven-day availability because those values exist in a copied template. Likewise, an old hospital affiliation should not remain current in structured data after the visible page is corrected.

Keep values that change frequently near their operational source. The clinic may publish a consultation fee, but a procedure estimate can depend on scope. Do not force a variable estimate into one unqualified price solely to populate a field. Explain the relevant boundary in visible text first.

Only use identity links that actually identify the same entity. A generic hospital home page, a medical council search form or a nearby clinic with a similar name is not automatically an equivalent identity page. Record why each link belongs.

Preserve identity when a doctor changes location or URL

Consider a fictional physician, Dr A, who works at Clinic North and Clinic South. There are three underlying professional or facility entities: one person and two locations. The pages describing those entities are an additional layer. A new page URL does not mean a new doctor has come into existence.

Choose stable public identifiers and maintain their relationships when a profile slug changes. The developer can map an old page URL to its new destination while retaining the intended entity identity. Do not use a patient record identifier, private account token or personal email address as a public schema identifier.

If the physician stops attending Clinic South, update the visible affiliation and location-specific service information. Do not delete Clinic South's identity if the clinic continues operating with other doctors. The change is a relationship and availability change, not necessarily the disappearance of every associated entity.

Test the public output on each relevant page after a change. It is possible for the main profile to be correct while a city listing or cached page still emits an old address. A consistent source of approved facts reduces that risk but does not remove the need to inspect the rendered result.

Give the developer an entity specification, not a keyword list

The following outline can be used before any code is written. It deliberately asks for factual ownership and relationships rather than promising a particular rich result. Use fictional data in a demonstration; do not paste real patient or private staff records into public testing tools.

A useful review includes the initial HTML, the rendered page and the structured-data block. Confirm that important information is readable without opening a hidden panel or relying on a marketing animation. Validation warnings should be interpreted against the feature being used instead of adding invented information just to make every warning disappear.

Retain the reviewed output and change date. When a field changes later, the team can compare the new assertion with the previously approved one rather than rebuilding the identity model from memory.

Make the content useful before asking an answer engine to cite it

Answer a patient's practical questions in ordinary text: who the doctor is, where the consultation takes place, which services are offered, which languages are available and how the booking is confirmed. A short, accurate answer with its qualifications nearby is easier to reuse responsibly than a long promotional paragraph containing several contradictory claims.

Google states that its AI search features use the existing SEO foundations and do not require special AI markup. It also does not guarantee inclusion. That guidance is Google-specific; it does not establish the selection rules of every assistant or answer engine.

Measure separate outcomes after an authorised publication: whether the intended URL is indexed, whether a relevant search brings a visit and whether a user completes a meaningful booking action. Keep those measures separate from schema validation. More properties in a JSON-LD block are not an outcome metric.

In a clinic software demonstration, ask to correct a doctor's language or location and inspect both visible text and metadata. That is a concrete consistency test. It does not establish that Bractis or any other platform currently supports every schema feature described here.

Source: Google Search Central: AI features and your website

Fictional example: two branches should not become two doctors

A physician's biography is shown on a main profile and on pages for two clinic branches. The website team keeps one professional identity and two facility identities, with branch-specific addresses and schedules. When the physician leaves one branch, that relationship is updated. The remaining branch and the physician's professional biography continue to describe the same entities; no new credentials or search-performance claim is created.

Map the public entities in your clinic website

Use these prompts in your clinic’s own notes. Record the result, responsible person and next action for each item.

Entity map
List the people, facilities and public pages separately. Which facts belong to each, and which relationships can change?
Claim evidence
For every credential, language, address and service, record the owner, evidence and last factual check.
Feature eligibility
Identify the specific Google feature, if any, and explain why the page meets its published scope.
Change rehearsal
Change one fictional location or language and compare the page, related listings and structured data for consistency.

Questions doctors ask

Will adding doctor schema improve rankings automatically?

No automatic ranking or AI-citation outcome is established. Structured data can describe content, but accuracy, page usefulness, feature eligibility and search-system decisions remain separate. Validate the markup and measure actual outcomes after publication.

Should every doctor directory listing use ProfilePage for Google?

Assess the actual page against Google's feature guidance. It focuses on a single affiliated person or organisation and has specific use cases. A listing or review page does not qualify merely because it is called a profile.

How should one doctor working at two clinics be represented?

Keep the professional identity separate from the two facility identities and describe the current relationships. Each location can have its own address and relevant service information. Do not duplicate or contradict the doctor's credentials across branch records.

Can markup contain qualifications that are absent from the biography?

Do not use hidden markup to make unsupported or stronger claims. Establish the qualification with appropriate evidence and present it accurately to readers before describing it in structured data.

Is a special AI file or schema required for Google's AI answers?

Google's cited guidance says there are no additional special technical requirements for its AI search features beyond the relevant search foundations. Eligibility still does not guarantee inclusion, and other engines may operate differently.

Sources and scope

Updated on . Technical and editorial guidance checked 25 October 2026. Vocabulary and search-feature rules are distinct; no live indexing, ranking, AI citation or product capability is claimed. No independent clinical or legal review is claimed.

Try the workflow
with Bractis.

Use a fictional doctor at two locations. Inspect identity relationships, visible claims and structured data after changing one branch schedule.

Request a clinic demo