Short answer: ChatGPT cites a page when it can find a direct answer to the question on it and can confirm who is behind it. For a practice that means answer-first pages, FAQ sections with matching schema markup, an llms.txt file, consistent directory listings, comparison content, and named, real results. When we rebuilt our own med spa page that way, ChatGPT cited it about two hours later. Here is exactly what changed, what the numbers were, and what they do not prove.
Most AEO advice is a list of tactics with no evidence attached. I would rather show you one real test with the honest numbers, including the parts that did not move yet.
What did we change?
On September 25, 2026 I rebuilt Northview’s med spa page around two questions we wanted to be named for: “AI visibility for med spas” and “AEO agency for med spas”. The changes:
- Headline and page title rewritten around the exact phrase a practice owner would use, instead of a generic outcome line.
- A nine question FAQ answering what owners actually ask, with FAQPage schema that matches the visible text word for word.
- A “What’s included” section that states plainly what the work covers, so an assistant has something concrete to quote.
- Named client results from real practices, not example numbers.
- An author byline on the page and in the schema, so the page has a named person behind it.
- Supporting guides published the same day and linked from the page: how to get a med spa recommended by ChatGPT, AEO vs SEO for med spas and how to choose an AI visibility agency.
- llms.txt updated with the new guides and a client results section. llms.txt is a plain text file at the root of a website that tells AI assistants what the business is and which pages matter.
What happened?
We track a fixed set of 12 med spa questions across ChatGPT, Perplexity and Google AI Overviews every week. In the September 25 run, about two hours after the rebuild, ChatGPT answered “Which marketing agencies specialize in AI visibility for medical spas?” by naming Northview Agency and citing northviewagency.com/med-spas/ as a source.
Now the honest part. That was 1 of 36 answers in that run. Perplexity named us 0 times and Google AI Overviews 0 times. One citation in one run is a start, not a trend, and I cannot prove the rebuild alone caused it; ChatGPT may simply have searched the web live and found a freshly updated page that answered the question directly. What it does show is that a page built to answer a specific question can be picked up fast.
What did the same work do for a medical practice?
Dr. Harrison’s Center of Functional Medicine in Denver went from 3 to 15 discovery calls a month. Northview runs AI visibility and patient acquisition for the practice, so more prospective patients find Dr. Harrison through search and AI answers and book a discovery call. I am not going to claim every one of those calls came from ChatGPT. The point is that the same system, applied to a real practice over time, turned into booked calls, which is the number a practice owner actually cares about. Visibility creates calls; a missed call system makes sure they get answered.
What does AEO work for a practice actually involve?
These are the pieces we build for every practice, and why each one matters to an AI assistant:
- Schema markup. Structured data stating the practice type, address, providers, services and FAQs in a format machines read directly.
- FAQ pages. Real patient questions answered in the first sentence, with schema that matches the visible text exactly.
- llms.txt. A short, plain text summary of who the practice is, who it serves and which pages to read first.
- Directory listings. The same name, address, phone and services on Google, Bing, Apple Maps and the directories in your specialty, because assistants cross check.
- Comparison content. Pages that answer “X vs Y” questions honestly, since patients and owners ask assistants to compare options, like this AI phone agent vs answering service comparison.
- Real, named proof. Results with a practice name attached. Unnamed case studies are easy to ignore.
One thing our tracking made clear: for med spa questions, the sources ChatGPT and the others cited most were third party lists, directories, an industry association site, LinkedIn and Reddit, not agency websites. Your own site is necessary, but the outside mentions are what confirm it. That is why directory and listing work sits next to the on-page work, not after it.
How can a practice tell if it is working?
- Write down 12 to 40 questions your patients actually ask, and lock the list.
- Run them in ChatGPT, Perplexity and Google AI Overviews and record who gets named and which sites get cited.
- Make the changes above, starting with the pages for the services that matter most.
- Rerun the same questions every week or month and compare. If the list changes, the numbers mean nothing.
- Watch the business number too: calls, consults and discovery calls booked.
What we are doing next
Our next step on our own site is the off-page side: listings on the directories that assistants cite for med spa questions and earning mentions on the lists they draw from. I will report the numbers as they come in, including if they stay flat.
If you want this run for your practice, here is how Northview’s AEO and SEO service works. If you are a functional medicine clinic like Dr. Harrison’s, see AI visibility for functional medicine.
Frequently asked questions
How long does it take for ChatGPT to cite a page?
It varies. Our rebuilt med spa page was cited about two hours after the changes went live, but that was one answer out of 36 in that run. Treat any single fast result as a signal, and measure a fixed question list over months.
What is llms.txt and does my practice need one?
llms.txt is a plain text file at the root of a website that summarizes the business and points AI assistants to its most important pages. It is quick to add and costs nothing. It is one signal among several, not a shortcut.
Can Northview guarantee my practice will be cited by ChatGPT?
No. Nobody controls what ChatGPT says, and anyone promising guaranteed placement is overselling. What we can do is build the pages, schema, listings and mentions assistants look for and report honestly how often your practice is named.
Does getting cited by ChatGPT actually bring in patients?
A citation only matters if it leads to calls. That is why we track business results alongside visibility. At Dr. Harrison’s Center of Functional Medicine, the combined work took discovery calls from 3 to 15 a month.
Is AEO different from SEO for a medical practice?
They overlap. SEO ranks your site in Google’s links, AEO gets your practice named inside AI answers. Most AEO work, like clear service pages and schema, also helps Google rankings, so we run them as one service.