BLOCKERS Episode 2. What should I do first?

BLOCKERS — a weekly series on removing obstacles to success in your practice.

I released the first issue of BLOCKERS last week, a series designed to help the thousands of health and wellness practitioners I work with get unblocked.

It was one of the most popular articles I’ve ever written.

What I didn’t mention is that last week’s article took me under an hour to create, because I wrote it with AI. This one too.

I’m telling you this now because I want to be honest with you about the process so you can leverage it. And because of what it says about the hour I spent, not about AI.

Stay with me for a moment while I show you how easy it is to figure out what you should do next to get your practice unstuck — or make your next million dollars.

For me, writing has never been a slow or difficult process. I love to write. It comes naturally to me.

The slow part, in almost every project I’ve personally stalled on in my career, always relates to not knowing where to begin, which end of the string to pick up first. And it always feels like a loop — like a knot you can’t figure out.

I’m terrible at knots.

And that’s the part I want to talk to you about today, because I’ve watched too many practices lose a year or more by being blocked by this — or actually close, because they can’t figure out how to start, maintain and scale a successful healthcare practice.

The loop 🪢

Here’s the loop, and tell me if you recognize yourself in it.

You know you should have a service offering, or better ones — that you can verbalize clearly enough to market and sell on your website and your social media and in your practice.

And it doesn’t have to be a content program. More often it’s an actual hands-on care program that takes place in your office, but it needs bones to be able to say what it is.

In some cases, that description of what you do could also be turned into an educational program that you could sell to people. Maybe a program you can charge for, or one you include in the care offering as patient education. But I digress.

The blocker for most people is that you can’t write the details of that care program, or the details of that patient education, until you know what the program actually is.

And you can’t define the program until you decide what’s included.

And you can’t decide what’s included until you know what you’d charge (or should charge, or can charge) — so that you know whether you have the time and the ability to make good on the offer in the first place. And you can’t price it until you know what it is.

So you’re stuck in a loop. You’re tied up in a knot.

So your growth sits. Not for a week — sometimes a year, sometimes three years.

And the way you describe it to yourself is:

I’m not a writer.

I don’t have time to write.

I didn’t go to school to learn how to describe what I do.

I can just hire a web person to make my web pages for me, or a social media expert to do social media for me.

But the reality is that if you don’t know what you’re supposed to be describing — in enough detail to hand off to somebody else — you’re still blocked. Because nobody else is going to write about what you do better than you can.

And to be quite frank, you are a writer. You write encounter notes. You write prescriptions. You give people care plans. And you talk about what you do to friends and family all the time.

So you’re not blocked on the ability to write it out. You’re blocked on what to do first. And not knowing what to do first feels like not having time. And you feel stupid sometimes. Right?

What makes this different from every other year you didn’t do it

For most of my career, the honest advice here was: hire someone, or block out a weekend and just do it. Both are real answers and both fail for the same reason — they cost money or a weekend, so you postpone them, and postponing is free.

That changed, and I don’t think most practices have absorbed how completely it’s changed.

You can sit down right now and open ChatGPT, or Claude, or Gemini, or Grok — pick one, it does not matter which, they’re all free — and have a conversation with an AI agent that will unstick you in 20 minutes. 100%. Guaranteed. Take that to the bank.

It costs nothing.

It takes twenty minutes.

You don’t have to install anything, hire anyone, or even tell your staff you’re doing it.

There is no version of this where “I couldn’t afford it” or “I didn’t have time” is true anymore. This blocker I’m talking about — where you don’t know what to do first — is officially gone in the age of AI.

And let me remove another small blocker in this category. Many of you have been embarrassed to tell another person that you don’t know what to do.

AI won’t judge you.

Most people use AI wrong on the first try and think it’s not for them

Here’s why plenty of doctors have already tried this and concluded it’s useless.

They typed something like this in ChatGPT:

“write me a blog post about weight loss” or “write me a care plan for someone who’s trending toward diabetes”

They got back garbage content that sounds like it could have come from any random website, and they moved on without using the tool correctly. That’s a reasonable reaction to a bad output. But it was the wrong input.

Garbage in. Garbage out.

The request that works is closer to this:

I run a functional medicine practice in South Florida. Most of my patients come to me for weight and metabolic issues. I have about twelve of them right now who’d probably pay for something structured, but I don’t have a program to sell them. I don’t know what to build first. Ask me whatever you need to know, then tell me what to do first.

That last sentence is the whole trick. Ask me whatever you need to know.

Now it’s a conversation with AI instead of treating AI like a vending machine. It will ask what you actually do with those patients today.

What you’re comfortable charging.

Whether you want it to run four weeks or twelve.

And somewhere in answering those questions out loud — which you have never once done, because nobody has ever sat you down and asked — you’ll find you already know most of it.

The output you get from the AI isn’t the point. The interrogation is the point.

You’re going to have an otherwise uncomfortable conversation with a robot that won’t make you feel uncomfortable, or stupid, or ask you why you don’t already know these things.

It’s literally peaceful and liberating.

This time, you’re the patient

You already know why the interrogation works, because you run it every day from the other chair.

A patient sits down and says I’m just tired. That’s not a diagnosis and it isn’t even really a complaint you can do anything with. So you take a history.

When did it start?

What changed?

What have you already tried?

What are you not telling me?

Twenty questions later you have something to work with — and half the time the patient hears their own answer come out of their mouth before you say anything.

That’s the same thing happening here. You’re describing a vague situation to a robot agent that will keep asking until your answers aren’t vague anymore.

The only difference is which chair you’re in.

That’s uncomfortable for people like you. You’re the one who asks the questions. You’re the one everybody else comes to. Sitting on the other side of that — admitting out loud that you don’t know what you’re supposed to do next — is a genuinely unfamiliar position.

Which is exactly why it’s easier to do with a machine at 10pm than with a consultant in a conference room, or worse, lying awake at night, spinning in your head.

And the same rule applies as with any history you take from a patient: the answer is only as good as what you put into it. The AI knows nothing about your practice except what you tell it, and it will confidently fill in any gap you leave.

So tell it what you actually charge, who actually walks through your door, what you’ve already tried, and what you flat-out refuse to do. The more specific you are, the less generic the answer.

And if you’re just starting and have no patients or offering, you can say that too.

What “first” usually turns out to be

I’ll spoil the ending for a lot of you, because I’ve had this conversation with enough practices to know how it goes.

The first thing you’ll end up creating is almost never a blog post or a piece of social media or an old school flyer announcing what you do.

Something else is blocking that.

The blocker is usually deciding on one program offering, for one kind of patient you already treat, at one price. Everything else — the content, the intake form, the website page, the email — falls out of that decision easily. It was never hard. It was just blocked by a decision you had not made yet, because you had no one to talk it out with.

That’s what made this loop so expensive. You spend a year not marketing yourself enough or not at all, or not offering anything you can market at all, because what you were actually not doing was making one choice that takes twenty minutes with someone who’ll ask you the right questions.

What to do today

Open ChatGPT, or Claude, or whichever one you like. Spend twenty minutes. That’s the only ask I have today.

Just pick one. Any one. Don’t research which is best — that’s another loop, and I can see you eyeing it. That’s a trap.

Tell it your actual situation in one paragraph. Real numbers, real patient types, real prices, real challenges. Not the version where the practice sounds good.

Ask what to do first — and tell it to interview you before it answers.

Then argue with it. When it suggests something that won’t work in your practice, say so and say why. That’s the most valuable part of the whole exercise, because explaining why something won’t work is how you find out what will.

Do the first thing it says before you close the tab. Even if you do it badly. Especially badly. Bad tries beat no tries because at worst, you learn.

You’ll come out with a decision, and the decision is the blocker. Everything after it is just typing (or dictating into chat, which then types for you 😉).

And where I come in

Nothing above requires John from BodySite. That’s not why I wrote this. I’d rather you go talk to chat than wait to talk to me about it, because you’ll come to me not knowing what to do.

But if you get to the end of that twenty minutes with chat and you decide on a care program or offering but no way to deliver it — that’s the next blocker. Great. That’s the one I’m actually useful for.

We have content libraries, we have a content delivery system, a way to manage and track patients through your offering. And a team of people who work with practices all day to get this type of thing up and running. If we don’t have what you need out of the box, we’ll build it with you. Forms, notes, plans, lessons. If you’d rather not do the setup yourself, we’ll do the setup.

Fair warning, though. I’d be doing you no favors by taking that work off your plate before you’ve made the decision, because the decision is the part nobody can do for you. Not me, and not the AI.

The AI just makes it take twenty minutes instead of another year.

Send me an email and say help. I’ll help you figure out what to do first.

John Cummings
Founder, BodySite.com

P.S. — I made a video that goes with this article using AI too. Seemed like the honest way to make the point. I’ll publish that later, and we’ll talk in Episode 3 about how easily you can remove the “how do I create video content” block.