There is a test I run with every member of our Growth Lab community. I ask them to open ChatGPT and search for themselves the way a patient would at 2am, using their problem in plain language, not a name, not a credential. Just: the thing they need help with, and the city they are in.


Most practitioners have never done it. And when they do, one of two things happens. Either they show up and they are genuinely surprised. Or they do not show up at all, despite years of investment in their website and their SEO, and they sit with that for a moment.


Both reactions are useful. Because this is the blood work of your marketing. It tells you exactly what is working and exactly what is not, faster than any report your SEO provider has ever sent you.


If you want to understand why search has changed and what the new landscape looks like, we covered that in detail here: SEO for Health Practitioners Has Changed.


This post is the next step. You know the game has changed. Now here is how to actually play it.


▶ Watch: Growth Lab Discoverability Sprint, Call 1 — watch what this looks like inside the room

01

Run the Test First. Everything Else Follows.

Before you change a single thing on your website, before you cancel your SEO provider, before you do anything, you need to know where you actually stand on the platforms your patients are actually using.


Right now, ChatGPT is fielding 42 million health inquiries every single day. Patients are not typing 'naturopath Vancouver' into Google and scanning a list. They are describing their situation in full sentences and asking for a recommendation. 'Who is the best concussion clinic in Toronto for adults?' 'I have been dealing with unexplained fatigue for two years. Who in Calgary actually specializes in this?'


The AI answering that question is not pulling from your keyword rankings. It is evaluating the entire body of digital evidence around you and deciding whether you are the credible expert in that specific arena. And it does this in seconds.


Your patients are not searching for you. They are searching for their problem. The question is whether you are the answer that comes up.

So before anything else: run the test. Open ChatGPT. Type in the search your ideal patient would run when they have finally hit their limit. Not your name. Their problem, in the language they would actually use. See what comes up. Then run the same search on Claude and on Perplexity, because the algorithms are different and the results will vary.


Record what you find. That is your baseline. Everything you do from here is about closing the gap between where you are and where you need to be.

THREE THINGS TO NOTE WHEN YOU RUN THE TEST

Do you appear at all? If yes, are you named directly or buried in a long list without prominence? And when you search using your city plus your specialty, versus searching by the patient's problem without using your specialty at all, do the results change? That last comparison tells you a lot about where your authority gaps actually are.

02

Understand What Is Probably Blocking You

If you ran the test and you did not show up, or you showed up inconsistently, it is almost never one thing. It is a cluster of signals that these platforms are reading as: this person is not actively playing.


I call this anti-marketing. Not doing nothing. Doing things that are actively working against your visibility without you realising it. And it is more common than most clinic owners want to admit, because it tends to live in the infrastructure nobody looks at.


The words on your site get you found. They are not what make someone stay. Most clinic owners are losing on both.


Here is what anti-marketing looks like in practice. Your site loads slowly on mobile, so the LLM scraping it reads a low-authority signal before it even gets to your content. Your privacy policy is a downloadable PDF with a date from three years ago, which tells these platforms you are not maintaining your digital presence.


Your last blog post was published in 2021. You are positioned as an expert in eight different things, which means the algorithm cannot position you as an authority in anything. Your Instagram bio describes you completely differently from your website headline.


It seems the passage you want rephrased is empty. Could you please provide the text youd like me to rephrase?


These platforms can scrape all of that in under two seconds. And every one of those signals factors into whether you get recommended or not.


The other layer is what happens after someone does find you. Even if you fix every technical signal, if a patient lands on your site and their gut reaction is that something looks dated or off-brand, they are gone. People make that determination in under three seconds. So anti-marketing is not just a back-end problem. It is also a first-impression problem that lives right on your homepage.

QUICK ANTI-MARKETING AUDIT

Load your site on your phone on a strong signal. Does it feel instant or are you waiting? Go to the footer. Is your privacy policy a clickable live page or a PDF? When was it last updated? Open your About page. Can you tell within two sentences what you specialise in? Now open Instagram. Does your bio say the same thing your website does? If any of these feel uncertain, you have found your starting point.

03

Build the Signals That Actually Get You Recommended

Once you know you are not showing up, and you have cleaned up the things actively working against you, the question becomes: what do you build?


The answer is not more content. It is not a keyword list. It is what I call an arena of authority. A specific, consistent, credible body of evidence across your entire digital ecosystem that points to one thing you are the best at.


LLM platforms are not trying to rank a page. They are trying to answer a question with the most trusted source available. The way they determine trust is by looking at whether the evidence around you, your website, your published content, your social presence, any mentions or citations on other sites, all consistently reinforces the same specific expertise. The more consistent and the more specific, the more likely you are to get recommended.


The tactics platforms use to index content will keep changing. Building authority in a specific arena is the one strategy you cannot lose.


This is also where the sunk cost fallacy tends to show up. Many clinic owners keep investing in infrastructure that is not working because they have already spent so much on it. That is not a strategic decision. It is a way of making yourself feel better about a previous investment. The question is not how much you have spent. It is whether what you have built is actually getting you found.


The practitioners who are showing up consistently on LLMs are not doing more. They are doing something more coherent. Their website, their content, their social presence, and their positioning all say the same thing. They publish regularly enough that AI models read them as active and engaged. And they have stopped trying to be the expert in everything, which means these platforms can confidently make them the expert in something.


  • Pick one arena. Not five. The condition, population, or outcome you want to be the go-to for. Everything you publish should reinforce this.

  • Publish something original at least monthly. A blog, a case study, a FAQ page that answers questions in your patients' language. Recency matters.

  • Audit your consistency. Does your website headline match your Instagram bio match your Google Business description match your LinkedIn headline? Inconsistency across platforms reads as low authority.

  • Check your technical foundations. Fast load speed, live privacy policy, structured About page, no broken links. These are table stakes for LLM visibility.

  • Ask your mom to search for you. Seriously. She will use the language a real patient uses, not clinical terminology. If she cannot find you, your patients probably cannot either.

Get the full searchability prompt bank

We built a set of copy-and-paste queries organized across three tiers of AI visibility, from confirming AI knows you exist through to the gold-standard test most practitioners have never run. Download it and work through all three tiers.


Download the FREE AI Visibility Prompt Bank →

There is no point picking curtains when the windows are leaking air. The clinic owners who are winning on LLM search right now are not the ones with the biggest marketing budgets. They are the ones who ran the test, found the gaps, cleaned up the anti-marketing, and built something coherent in a specific arena.


Start with the test. Record your results. Find the first thing that needs fixing. Then fix that before you add anything new on top.


Inside Growth Lab we spent six weeks on this, moving from the foundational audit all the way through to the operational systems that make discoverability a built-in part of how you show up and publish. It is not a one-time project. It is a way of running your practice.

Growth Lab is where clinic owners build practices that get found, get booked, and keep growing without them running every lever. The searchability sprint is one piece of a much bigger system.

Frequently Asked Questions

How do I know if I am showing up on ChatGPT, Claude, or Perplexity?

Open each platform and search the way your ideal patient would, using their problem and their city, not your name. Record whether you appear, and whether you are named directly or buried in a list. Run this test monthly. It changes faster than any SEO report.


My SEO provider says my rankings are fine. Why am I not showing up on AI platforms?

Google rankings and LLM visibility measure different things. LLMs evaluate your whole digital ecosystem, not just your position in search results. A slow site, stale content, inconsistent messaging across platforms, and overly broad positioning can all block you even if your Google rankings look healthy.


What is anti-marketing and how does it affect my LLM visibility?

Certainly! Heres a rephrased version of the passage: Anti-marketing occurs when your current setup inadvertently undermines your efforts. Examples include a privacy policy PDF showing an outdated date, a website that loads slowly on mobile devices, social media profiles that conflict with your website information, and content that hasn’t been refreshed in years. These factors quickly signal to AI platforms that you lack credibility and activity as an authority. Such signals are collected instantly and carry significant weight.


Do I need to create a lot of new content to show up on AI search?

Not necessarily. Consistency and specificity matter more than volume. Publishing one original, well-structured piece per month in your specific arena of expertise will outperform a high volume of generic content. Fix the anti-marketing gaps first, then build from there.


What is the single most important thing I can do right now to improve my LLM visibility?

Run the test. Open ChatGPT and search the way your patient would. That tells you exactly where you stand and what needs to happen first. Everything else is built from that baseline.