AI visibility for practices: what AI assistants say about you

What AI visibility means for a local practice

For a software company, AI visibility means share of voice across thousands of product questions. For a practice it is narrower and more practical: a handful of questions, asked by people within driving distance, that end in a phone call or a booking.

Three things are worth tracking:

• Being named. Does the assistant list your practice when someone asks for your specialty in your town?
• Being described correctly. Are your hours, insurance plans, services and address right? An assistant that names you but says you aren't taking new patients does real harm.
• Being cited. Does the answer link to your website, or only to a directory page about you?

The search demand around the term itself is modest. Most of it comes from marketers trying to understand the idea.

Google Ads Keyword Planner data via DataForSEO, US, 12-month average to Sep 2026. "ai visibility" has a DataForSEO keyword difficulty of 26.

Most pages ranking for these terms are written for brand marketers and software buyers (DataForSEO results sample, Oct 2026). This section is written for practices, so every example below is a patient asking about care.

Want a baseline before reading further? The free marketing audit checks whether Claude and Gemini name your practice and whether AI crawlers can read your site.

How patients use AI assistants to choose a practice

Be careful with claims here, including your own. The best recent national number is about health information, not choosing a doctor. In a KFF poll (opens in a new tab) of 1,343 US adults, fielded February 24 to March 2, 2026, 32% said they had turned to AI chatbots for health information in the past year. Among those users, 41% wanted to look something up before deciding whether to see a provider.

We have not found a national survey that measures how many patients pick a practice through an assistant. Your own analytics are better evidence. Visits that arrive from chatgpt.com, perplexity.ai, gemini.google.com or claude.ai usually show up as referrals, and the trend line tells you more about your patients than any industry estimate.

The questions themselves fall into four patterns. Illustrative examples, all fictional:

Discovery questions decide whether you're considered at all. Verification questions decide whether a patient who already found you keeps going. Both depend on the assistant having accurate, current facts about you.

Where assistants get their information

Assistants answer from two places: what the model learned in training, and what it fetches from the web when the question needs current or local facts. A question like "best dermatologist in Plano" almost always triggers a live search, so the live sources matter most for a practice.

Here is how each provider documents its own approach, as of the date this page was checked:

For a local practice, the places an assistant can find facts about you fall into four groups: your own website, your Google Business Profile and other listings, review platforms, and directories or health-system pages that mention you. Google is explicit about one of them: its guide to generative AI features says keeping Business Profile information current can help your services appear in AI responses (Google Search Central (opens in a new tab)). The other providers don't publish which local sources they prefer.

GEO, AEO and SEO: how they relate

Three labels describe overlapping work:

• SEO (search engine optimization) aims to rank your pages in search results.
• GEO (generative engine optimization) aims to get your content used and cited inside AI-generated answers. The term comes from a 2023 research paper.
• AEO (answer engine optimization) focuses on writing short, direct answers to the specific questions people ask.

Google's position is that these are not separate disciplines for its own products. Its generative AI guide says optimizing for generative AI search "is optimizing for the search experience, and thus still SEO." For a practice, the useful takeaway is that the foundations are shared: a crawlable site, a page per service, accurate listings, and answers a patient can trust. The guides at the end of this page cover each label in depth.

How to check your AI visibility, manually and weekly

You can audit this yourself in under an hour a week.

1. Write five to ten patient prompts. Mix specialty plus city ("family doctor in Boise"), service plus city ("IUD insertion Boise"), insurance ("pediatrician Boise that takes Blue Cross"), and your own name.
2. Use a clean session. Log out or use a private window, and write down the city you're testing from. Assistants infer location, so a test from home and one from the office can differ.
3. Run every prompt in every assistant. Record who is named, in what order, which pages are cited, and any wrong facts about you.
4. Repeat on the same day each week. One run is a snapshot. Four weeks of runs show a pattern.
5. Check crawler access. Open yoursite.com/robots.txt and confirm you aren't blocking OAI-SearchBot, PerplexityBot, Claude-SearchBot or Googlebot.

A simple log is enough:

For a quick first read, the free AI visibility checker runs a single check. When you want to compare paid options, our guide to AI visibility tools explains what each kind of tool measures.

How Medfy checks it

Medfy's AI visibility tracking runs weekly checks of ChatGPT, Gemini, Claude, Perplexity and Google AI Overviews using the questions patients ask, such as "best family doctor near me" in your city. It shows whether you're named, who is recommended instead and how that changes. Medfy's first-party analytics counts visits from AI assistants, so you can see whether mentions turn into traffic.

Before subscribing, the free marketing audit gives an AI visibility score: it asks Claude and Gemini, with live web search, whether they name your practice, and checks that the crawlers behind ChatGPT, Perplexity, Claude and Gemini can read your site. Tracking is part of the base plan; see Medfy pricing.

What improves your odds of being mentioned

None of these steps can make an assistant name you. Each one makes you easier to find and easier to verify.

• Answer the question on your own site. One page per service, with plain facts a patient needs: who it's for, insurance accepted, new-patient availability, what the first visit involves. Assistants can't cite what isn't written down.
• Keep your listings complete and identical. Same practice name, address, phone and hours on your Google Business Profile for doctors, Bing Places, Apple Business Connect and the directories patients use.
• Earn reviews steadily. Google says more reviews and positive ratings can help local ranking (Google Business Profile Help (opens in a new tab)), and its AI features draw on the same local information.
• Let search crawlers in, separately from training crawlers. OpenAI notes that each crawler setting is independent, so you can allow OAI-SearchBot while blocking GPTBot. Anthropic warns that blocking Claude-SearchBot may reduce your visibility in Claude's search results.
• Source your health content. Link clinical statements to MedlinePlus, PubMed or your specialty body, and show who reviewed the page and when.
• Watch Google's AI features too. They behave differently from chat assistants; our guide to Google AI Overviews covers them.

What you can't control

Being honest about limits saves money.

• Answers vary. The same question can produce different practices depending on phrasing, the patient's location, the day and the model version. Google's own help page says AI Overviews "can and will make mistakes" (Google Search Help (opens in a new tab)).
• Personalization. Google's AI Mode can tailor responses using previous searches for signed-in adults who have Search history turned on (Google Search Help (opens in a new tab)). Your test and your patient's view may not match.
• No paid placement in answers. Google states there is no way to request or pay for a better local ranking. No provider documents a way to buy a mention inside an answer.
• Health caution. Medical questions pull answers toward official sources. In Pew Research Center's (opens in a new tab) March 2025 browsing study, Google's AI summaries linked to .gov sites in 6% of cases versus 2% for standard results. An assistant may also tell the patient to see a clinician rather than name anyone.
• No insider tools. Google says no third-party tool has access to its internal ranking or AI systems. Tools observe answers; they don't control them.

No agency, tool or platform, Medfy included, can promise that an assistant will recommend your practice.

Guides in this section

• Generative engine optimization (GEO): what the original research found and which parts apply to a practice.
• GEO vs SEO: a side-by-side of goals, signals and measurement, and where to spend limited time.
• Answer engine optimization: turning the questions patients ask into short, sourced answers.
• How to rank in ChatGPT: four practical steps and a weekly test protocol, without hacks.
• How AI citations work: why health answers cite authorities and what your pages can be cited for.
• Perplexity for business: using Perplexity's visible sources to audit your presence.
• Local AI search: how assistants handle "near me" questions and why two patients see different answers.

Frequently asked questions

What is AI visibility?

AI visibility is how often, and how accurately, AI assistants mention your practice when patients ask for care. For a practice, that means checking whether ChatGPT, Gemini, Claude, Perplexity and Google AI Overviews name you for questions like "best family doctor near me" in your city, what they say about you, which sources they cite and who they recommend instead.

Do patients really use ChatGPT to find doctors?

Some do, but we have not found a national survey that measures how many choose a practice that way. KFF's March 2026 poll found 32% of US adults used AI chatbots for health information in the past year, and 41% of those users wanted information before deciding whether to see a provider. Your own analytics are better evidence: count visits from chatgpt.com, perplexity.ai, gemini.google.com and claude.ai.

How do I know if ChatGPT recommends my practice?

Ask it, the same way every week. Use five to ten patient-style prompts, such as "best pediatric dentist in Tulsa", in a logged-out session, and note the city you're testing from. Record whether you're named, in what position, which practices appear instead and which pages are cited. Expect answers to shift between runs; the pattern over several weeks is what counts.

Can I pay to be recommended by AI assistants?

No provider documents a way to buy a mention inside an assistant's answer. Google states there is no way to request or pay for a better local ranking, and its AI features draw on the same Search systems. OpenAI's documentation describes a separate system for ads on ChatGPT, and an ad is not the same as being named in the answer. Be wary of anyone selling mentions.

Is AI visibility different from SEO?

It overlaps heavily. Google says optimizing for its generative AI features is still SEO, because they rest on its core ranking systems. The differences are in measurement and sources: assistants summarize several pages, cite directories and reviews alongside your site, and give different answers to different people. Good SEO is the base; AI visibility adds a weekly check of what the answers actually say.

How long does it take to improve AI visibility?

Expect months, not days, and no certainty. Some technical fixes register quickly: OpenAI says robots.txt changes for its search crawler take about 24 hours. Content, listing and review changes depend on recrawling and on how each assistant weighs sources, which can take weeks or longer. Track the same prompts weekly so you can tell real movement from day-to-day noise.