SEO and AI search hub · Updated November 9, 2026

Healthcare SEO and AI Search: What Multi-Site Operators Should Actually Do

AI answers in Google, Bing and ChatGPT draw on the same web your patients search. This hub separates what Google and Microsoft officially say from what vendor studies report, and lays out who should own location data, provider pages and reviews when you run many sites.

What Google and Bing say What changed recently By the numbers Jump to the ownership workflow

Why this became a hub

This wasn’t supposed to be a hub. I started Healthcare Growth Strategies to keep the articles I wanted to come back to, and AI search was one topic among many. Then it stopped behaving like one topic. In a few weeks, Zocdoc opened its directory to AI booking agents, Google widened AI features across Search and Business Profiles, and screenshots surfaced of ChatGPT preparing to let businesses claim their listing. It’s changing too fast to leave scattered across the library, so this page is where I’m collecting it: dated, with what’s confirmed kept apart from what’s only reported.Paul Gruensfelder, November 2026

What changed, and what did not

AI search added new places for your brand to show up, or to show up wrong. It did not replace the basics. Pages still have to be crawlable, indexed and trustworthy, and every location still needs the right address, phone, hours and website everywhere a patient or an AI system looks.

For a multi-site operator, the hard part is rarely a clever tactic. It is keeping dozens or hundreds of location records, provider profiles and review responses accurate at the same time, with clear owners.

What changed recently

Newest first, updated weekly. “Reported” means seen or claimed, not announced. Vendor studies are labeled.

Nov 6, reported: ChatGPT is building a “Claim your business” hub with daily mentions and clicks for verified businesses. OpenAI has not announced it.
Nov 6, vendor study: Most multi-location brands get mixed AI results across their own locations, in SOCi’s data on about 3,000 brands.
Nov 6: AI Overviews cite under half of the top results on health queries.
Nov 4: AI citations churn day to day, so measure over a week, not a day.
Nov 3, vendor study: Reddit ranking threads correlate most with AI visibility across 10,000 brands. Correlation only.
Nov 3, vendor study: OpenAI publisher deals do not guarantee ChatGPT citations, and Reddit’s share of answers fell sharply.
Nov 1: AI Overviews now appear on most branded searches.
Oct 30: Google will support a shared robots.txt control for AI training in 2027. It covers training, not search visibility.
Oct 29, pilot: Google is widening an AI agent that answers Business Profile messages. Check what it may say to patients before turning it on.
Oct 29, vendor study: Reviews act as a filter for AI recommendations, not a ranking factor, in SOCi’s data.
Oct 28: Zocdoc lets AI agents search providers and book appointments. Your directory profile is now an AI data feed.
Oct 28, vendor guide: Why Google keeps changing your Business Profile, and why groups with 100+ listings get no edit emails.
Oct 22, vendor study: For dental questions, ChatGPT mostly cites practice websites. In medicine, hospital directories lead.
Oct 22: ChatGPT still sends under 1% of Google organic traffic. Keep it in proportion.
Sep 24, vendor study: ChatGPT now targets specific sites in most of its searches, which makes brand-topic association matter more.

Signals we’re watching

When ChatGPT’s Business Hub launches, and what proof of ownership it asks for.
OpenAI’s claim that ChatGPT is at about 20% of Google’s search volume. No method has been published.
Whether Google’s Business Profile messaging agent rolls out widely, and how it handles health questions.
Semrush stopped tracking Microsoft Copilot as of November 9. The reason has not been confirmed.
Whether other AI providers adopt the shared training control Google plans to support in 2027.
Healthcare-specific data on how patients find providers through AI. Little has been published this fall; it goes here when it is.

Inclusion of vendor research is not an endorsement by Healthcare Growth Strategies, Paul Gruensfelder or his employer.

Healthcare AI search by the numbers

The figures worth knowing, each with its source, sample and date. Vendor data is labeled; figures are as each source reports them and were not independently verified.

Figure	What was measured	Source and date	More
48.7%	Google health care queries showing an AI Overview, the highest of 10 industries; AI referrals were 0.64% of health care site traffic	Conductor benchmarks (vendor data), data May-Oct 2025, updated April 2026	Read the summary
44.4%	Share of top organic results that AI Overviews cited, across 3,611 health questions (AI Overviews appeared on 96.4% of them)	Northwestern University study, October 2026	Read the summary
82.9%	Branded searches showing an AI Overview by September 29, 2026	Ahrefs (vendor data), September 2026	Read the summary
0.62%	ChatGPT referral sessions as a share of Google organic sessions, 100 websites, 12 months of GA4 data	Marie Haynes analysis, October 2026	Read the summary
47%	Surveyed patients who had used an AI chatbot to research a provider; about two-thirds of those AI users ran into wrong addresses, phones, insurance or hours	rater8 2026 Patient Choice Report (vendor survey) via TechTarget, June 2026	Source
53.8%	ChatGPT citations that went to practice-owned websites on dental questions (13,152 citations, 20 metros); for physician questions, hospital directories led at 27.1%	Halcy study (vendor data) via Dental Economics, October 2026	Read the summary
18.6%	Locations of multi-location brands that never appeared in ChatGPT scans (16,240 scans, 1,762 brands)	Birdeye (vendor data), October 2026	Read the summary
1.2% vs 35.9%	Brand locations ChatGPT recommended, versus how often the same brands appeared in the Google 3-Pack	SOCi Local Visibility Index (vendor data), October 2026	Read the summary
About half	AI citations that disappeared the next day, of 37,500 tracked for 20 days; about 75% of dropped citations later returned	Ahrefs (vendor data), October 2026	Read the summary
About 30x	Growth in AI-agent traffic to Zocdoc since July, as reported by the company	Zocdoc via Healthcare Dive, October 28, 2026	Read the summary

Reviewed November 9, 2026. Studies use different samples and methods, so compare figures with care rather than side by side.

Official guidance: Google and Bing

These points come from the search engines’ own documentation. Treat them as the baseline before any vendor advice.

Eligibility is the same as regular search. Google says a page must be “indexed and eligible to be shown in Google Search with a snippet” to appear in its AI features, and that there are “no additional requirements to appear in AI Overviews or AI Mode.” (Google: AI features and your website, updated Dec. 10, 2025)

No special AI files or schema required. Google states you do not need new machine-readable files, AI text files or markup to appear in these features, and that there is no special schema.org structured data to add. Structured data is still listed as a good practice when it matches the visible text on the page. (same page)

Helpful content and accurate business data. Google’s guide on generative AI features recommends non-commodity content that is “helpful, reliable, and people-first,” good crawlability and page experience, and up-to-date Business Profile and Merchant Center information. It also says there is no requirement to break content into tiny pieces. (Google’s guide to optimizing for generative AI features, updated July 10, 2026)

Measurement. Google reports AI feature traffic inside the overall Web search type of the Search Console Performance report, not as a separate total. Bing launched an AI Performance report in Bing Webmaster Tools as a public preview on Feb. 10, 2026, and added a Topics Portfolio beta on Oct. 13, 2026. (Bing Webmaster Blog)

Controls. Google honors nosnippet, data-nosnippet, max-snippet and noindex for AI features, the same controls used for regular snippets.

What vendor and independent studies report

These studies are useful signals, not rules. Each one has a narrow sample, and several come from companies that sell AI visibility or SEO tools.

Locations can be invisible or wrong. Birdeye, which sells listings and AI search software, reports 16,240 location-level ChatGPT scans across 1,762 multi-location brands. In its data, 18.6% of locations were invisible, ChatGPT showed the wrong website for 57% of locations and the wrong phone number for 35%, and only 0.6% returned fully accurate information on every field it measured. Scan dates are not stated. HGS summary

ChatGPT searches less often than many assume. Ahrefs, which sells an AI visibility tool, analyzed 66.6 million free, logged-out ChatGPT responses from August and September 2026. 14.4% contained fan-out search queries, averaging about 1.6 queries each. In a separate test, factual questions cited an official source more often than comparison questions (84.4% vs. 68.3%). Ahrefs notes free sessions search less than paid or reasoning models. HGS summary

AI referrals favor different pages than organic search. A Search Engine Land contributor study of 10 hand-picked websites in March 2026 found the top 10 organic pages captured 55% of organic sessions but only 29% of LLM sessions. One month and ten sites cannot show a general rule. HGS summary

Figures above were re-checked against each source on October 31, 2026. They are the vendors’ or authors’ figures and were not independently verified by HGS.

Who owns what: a multi-site governance workflow

Many accuracy problems come from unclear ownership, not missing tactics. This is an HGS working model to adapt, not an industry standard.

Asset	Usual owner	What to check	Suggested cadence
Location data (Business Profile, listings, website location pages)	Marketing or digital team, with each site’s manager confirming changes	Name, address, phone, hours, website URL and services match across Google, Bing, Apple, directories and your own site	Monthly, and same week as any move, phone change or hours change
Provider pages and profiles	Marketing, fed by credentialing or the medical staff office	Credentials, specialties, locations, accepting-new-patients status; departed providers removed or redirected	With every hire, departure or schedule change
Reviews and responses	Site leaders respond; marketing sets policy and monitors	Response time, privacy-safe replies, recurring themes passed to operations	Weekly
Service and condition pages	Marketing, with a named clinical reviewer	Accurate, specific, original content; reviewer and review date visible	At least yearly, and when services change
Technical health	Web or SEO owner	Indexing, robots rules, snippet controls, redirects, structured data that matches visible text	Monthly
Measurement	Marketing analytics	Search Console, Bing Webmaster Tools (including AI Performance) and analytics referrals, read by location where possible	Monthly

Where to start

Can search engines see the pages? Confirm key location, provider and service pages are indexed and snippet-eligible. This is Google’s stated entry requirement for AI features.
Is every location’s basic data right? Spot-check a sample of sites in Google, Bing and an AI assistant. Wrong websites and phone numbers are the errors Birdeye reported most often.
Are provider pages current? Departed providers and stale accepting-new-patients flags send patients to dead ends.
Is the content worth citing? Replace generic service copy with specific, reviewed information patients cannot get everywhere else.
Are you measuring it? Set a monthly view of Search Console, Bing Webmaster Tools and AI referrals before judging any new tactic.

Go deeper: HGS resources

Limitations

AI search products change often. Official pages cited here were current on the review date below; check them again before major decisions.
Vendor studies use their own samples and methods, often without full disclosure, and many come from companies selling related tools.
Referral and citation counts show association with visibility, not proof that a tactic caused new patients.
Healthcare search carries extra trust expectations. Clinical content should be reviewed by qualified clinicians, and nothing here is legal or compliance advice.

Put it in your marketing plan

See how search visibility fits with reputation, access and measurement across a multi-site organization.

Read the healthcare marketing strategies guide