How to run a healthcare SEO audit (technical and content)

A healthcare SEO audit should answer three questions: can Google index the pages that matter, does the medical content deserve to rank, and is each page tied to a real unit, doctor and service. Run technical checks first, then content, then local and entity signals. End with a ranked fix plan grouped by template, with named owners, not a crawler export.

I have commissioned audits that arrived as two-hundred-page PDFs, with every warning from every crawler pasted in and no view on what mattered. The web team read the first ten pages and filed the rest. A healthcare SEO audit is only useful if it ends in a short, ranked list of fixes, each with an owner and a reason a patient would notice the difference.

This is the method I use for a hospital website audit, technical and content together. It works for a single-specialty hospital and for a group with thousands of doctor pages, because it is organised around page types rather than around tool output. It sits inside the wider local SEO playbook for hospitals, and it is the first thing I run on any new site.

What a healthcare SEO audit has to answer

Strip away the tooling and an audit answers three questions. Everything you check should map to one of them, and anything that does not can wait.

1. Can Google find, crawl and index the pages that matter? Doctor pages, specialty pages, unit pages and the condition and procedure pages that carry search demand.
2. Does the content deserve to rank for medical queries? Medical content is judged more strictly than most. Google’s guidance on creating helpful, reliable content says it gives more weight to strong E-E-A-T on topics that could affect people’s health, and that trust matters most.
3. Is each page tied to a real place, doctor and service? A cardiology page that never says where, when and with whom will not win a local query, however well written.

The technical strand answers the first question, the content strand the second, and the local and entity checks the third. I run them in that order because content fixes are wasted on pages Google cannot index.

Before you crawl: scope, access and a baseline

Most audits stall in week one waiting for access. Ask for all of it on day one, in writing, with a named person on the hospital side who can grant it.

• Google Search Console for every property, including old domains and subdomains still live.
• Analytics and tag manager, with read access at minimum.
• CMS access, or at least an export of all published URLs with page type and last-modified date.
• Server or CDN logs for a recent period, if the site is large.
• Business Profile manager access for every unit, department and doctor listing the hospital manages.
• The master doctor list from HR or medical administration, with units and timings.

Then build a page inventory by type: doctors, specialties, units, conditions and procedures, blogs, campaign landing pages, and everything else. The inventory matters more than any crawler report, because it tells you how big each template problem is. Finally, record a baseline of organic entrances by page type, profile actions by unit and enquiries by source, so the audit can be judged later.

Technical audit: crawl, index, render

Hospital sites have a few technical failure modes that I see again and again. Check these before the generic list.

Indexation of the pages that matter

Start with the Page indexing report in Search Console and filter by page type. The statuses to worry about on a hospital site are “Crawled – currently not indexed” and “Discovered – currently not indexed” on doctor and specialty pages, “Duplicate without user-selected canonical” on doctor pages repeated across units, and “Soft 404” on thin or empty profiles. Compare the count of indexed doctor pages with the count of doctors on the HR list. The gap is usually the first headline finding.

Crawl traps in the doctor finder

Find-a-doctor tools often generate a URL for every combination of specialty, unit, language and gender filter. Crawlers can spend their time on these combinations instead of on real pages. Google’s guide to managing crawl budget is written for very large sites, but its advice applies to any hospital directory: consolidate duplicates, keep low-value parameter URLs out of the crawl, and return 404 or 410 for pages that are permanently gone.

Rendering

If doctor lists, timings or booking widgets load only through JavaScript, check what Google actually sees using the URL Inspection tool’s rendered HTML. I have seen doctor pages where the name and specialty were visible to patients but absent from what the crawler rendered.

Status codes and redirects

Crawl the whole site and list every 404, every redirect chain, and every page that redirects to the homepage. Departed doctors are the usual source. Old campaign microsites and retired package pages come next.

Legacy domains and duplicate properties

Hospital groups accumulate domains: an old unit website from before an acquisition, a campaign microsite, a separate domain for health checks, plus the usual http, https, www and non-www variants. List every one, check whether it is live, redirecting or indexed, and make sure each old URL with links or traffic points permanently to its closest equivalent. Orphaned domains quietly compete with the main site for your own brand and doctor names.

Regional-language versions

If you publish Hindi, Marathi, Tamil or other language pages, check that each version is linked from its English equivalent, is not canonicalised to the English page, and declares its language alternates correctly. I often find language pages that exist in the CMS but are unreachable from navigation, which means neither patients nor crawlers find them.

Speed and stability

Core Web Vitals are LCP, INP and CLS, and web.dev sets the good thresholds at 2.5 seconds, 200 milliseconds and 0.1, measured at the 75th percentile of page loads. On hospital sites the usual culprits are hero sliders, chat widgets, multiple tag scripts and unoptimised doctor photos. Test the doctor, specialty and unit templates on a mid-range Android phone over mobile data, which is how most of your patients arrive. The detail is in Core Web Vitals for hospital websites.

Content audit: the medical website SEO checklist

The content strand is where hospital sites differ most from other sites. Run each page type through this medical website SEO checklist, sampling if the template is large.

• Named clinical reviewer and review date on every condition, procedure and treatment page.
• Clear authorship, with a byline that links to a real profile of the doctor or writer.
• No outcome promises or superlatives. “Best”, “guaranteed” and success-rate claims without a source are both a trust problem and a regulatory one. The marketer’s view of the rules is in what hospital ads can and cannot say.
• Duplication across units. The same doctor bio or specialty text copied to every unit page.
• Thin specialty pages. A heading, a stock photo and a paragraph of generic text, with no doctors, services, timings or unit named.
• Cannibalisation. Two or more pages targeting the same condition in the same city, often a blog post and a service page.
• Staleness. Pages that mention retired technology, departed doctors or expired offers.
• Language coverage. Whether the regional-language pages that exist are indexed, linked and correctly marked up.
• AI-drafted content without review. If the team uses AI to draft at scale, check the review trail. AI content at scale without wrecking medical accuracy sets out the controls.

Also check the pages a patient needs before trusting you: accreditation, pricing approach, privacy notice, grievance contact. What a hospital website must show lists them. How Google’s raters think about all of this is covered in YMYL and E-E-A-T for medical websites.

Local and entity signals

This strand checks whether the site and the listings describe the same hospital. It is quick, and the findings are usually embarrassing.

• Every unit page shows the same name, address and phone number as that unit’s Business Profile.
• Every Business Profile links to the matching unit page, not the homepage, with campaign tags so profile visits can be separated from other organic visits in analytics.
• Specialty pages link to the doctors who practise that specialty at that unit, and doctor pages link back.
• Structured data validates and matches what is visible on the page. The implementation detail is in medical schema markup.
• Doctor pages exist once, with one canonical URL, even when the doctor consults at several units. Doctor profile pages that rank covers the structure.

Scoring and prioritising findings

A flat list of issues is how audits die. I score every finding on patient impact and effort, then group by template, because fixing one template can fix a thousand pages. The table shows how typical findings tend to land; your scores will depend on your site.

Anything scored high impact and low effort goes into the next sprint. High impact and high effort becomes a project with a business case. Low impact items go into a backlog that nobody should feel guilty about.

Running the audit: who does what, and when

For a mid-sized hospital site, this is the sequence I follow. A large group will take longer on inventory and remediation, not on the method.

1. Kick-off and access. Agree scope, page types in scope and the named owners. Collect every access on the list above.
2. Inventory and baseline. Build the page inventory by type and record the baseline numbers.
3. Technical crawl and Search Console review. Run the crawl, pull indexation by page type, test templates on mobile.
4. Content sampling. Review a sample of each template against the checklist, with a clinician available for questions.
5. Local and entity checks. Compare unit pages, profiles and structured data line by line.
6. Scoring workshop. Marketing, web, medical administration and a clinical reviewer score findings together.
7. Fix plan. A one-page summary for leadership and a ticketed backlog for the web team, grouped by template.

If an agency is doing the audit, write this sequence into the scope and ask for the fix plan as the deliverable, not the crawl export. Insist that every finding names the page type, the number of URLs affected, the likely patient impact and a suggested owner. An agency that cannot say which findings it would fix first has not finished the audit. Also ask for the crawl configuration it used, so your own team can repeat the crawl later and compare like with like. For a checklist version to work through, use the healthcare local SEO audit checklist.

Keeping the healthcare SEO audit alive

A one-off audit decays fast on a hospital site, because doctors join and leave, units open and campaigns come and go. I turn the audit into three habits.

• Release checks. Any change to the doctor, specialty or unit template is tested for indexability, structured data and mobile speed before it ships.
• A monthly directory sync. The HR doctor list is reconciled against the website and the profiles, with joiners and leavers handled within an agreed time.
• A leaver protocol. When a doctor leaves, the web team gets the same notice as IT: update or redirect the page, remove the doctor from unit and specialty pages, and flag any practitioner listing.
• A quarterly re-run of the technical crawl and Search Console review, plus a content sample on one page type in rotation.

The patient never sees your audit. They see a doctor page that loads, a phone number that works and a specialty page that tells them where to go. Judge the audit on that, and on the booked appointments that follow.

Questions people ask