My Best Courses Were Total Duds

One of my courses did well over a million dollars. When I finished recording it, I thought it was okay at best. That was my honest review of my own product: okay at best.

And some of the best courses I ever made — the ones I was actually proud of, the ones I'd have pointed you to if you asked what my best work was — were total duds. Barely sold.

That's the ending of the story. I'm giving it to you first because if you're a professional sitting on a course idea — or two, or three — I know the stage you're in. You're comparing them. You're researching. You're waiting until you feel sure the idea will sell before you commit real time to building it. That wait feels like diligence. My own catalog says otherwise, and I'd rather you hear it now than three years from now.

The one that hit

The Elite Nurse Practitioner had over forty courses by the time I sold the business. I launched every one of them, and I had opinions about them the way you have opinions about your own work. I knew which ones were thorough. I knew which ones I'd rushed. I knew which ones I'd show off.

The market ignored every one of those opinions.

The course that did over a million was the "okay at best" one. It wasn't my most polished work — most of my courses went out the door about eighty percent complete, and I figured out the last twenty percent over the year after launch, mostly from what students asked. None of that slowed it down. Buyers weren't grading my craft. They had a problem in their career that hurt, the course made it stop hurting, and they told their colleagues.

The same pattern showed up outside the courses. My first website was junky — genuinely junky, and it stayed junky for about a year — and the whole business still did almost a million dollars in year one through that junky website. Buyers kept ignoring the things I would have fixed first.

The ones that died

Then there's the other column. Some of my best courses — best by the only scale I had, which was mine — went nowhere. The material was strong. The teaching was careful. I picked those subjects because I cared about them, which is exactly why they were my best work. The market shrugged at every one of them.

It stings for a while, and then it teaches you the most useful thing I know about this business: my judgment of my own work was measuring craft. The buyer was measuring pressure — how badly the problem in front of them hurt, and whether this thing would make it go away. Those are two different scales, and one doesn't translate into the other.

Now, I want to be precise about the lesson, because it isn't "quality doesn't matter." Ship something that doesn't deliver and you'll pay for it in refunds, and in a profession, word gets around. The lesson is narrower and stranger than that: past a reasonable bar, quality stops predicting anything. You just never know. And I say that as the person holding the sales data — years in, forty-plus launches in, I still couldn't have told you in advance which of my own courses would win.

What the tests can't tell you

You might push back here, because I teach idea tests myself. There are real questions to put to a course idea before you build anything — whether the problem is real, whether it's big enough to carry a business — and I've written about three of them. Run them. They work, and they're the part of this you can actually know in advance.

But watch what they tell you. The tests qualify an idea: this problem is real, a business can stand on it. What no test tells you is the ranking. Among the ideas that pass, which one does three times the revenue of the others? Which "okay" course embarrasses your favorite? I had live sales data on my own catalog and I still guessed wrong about my own products. You're trying to make the same call from a notebook.

That's the certainty you've been waiting to feel before you start. It's not coming. There's nothing out there that produces it.

What to do instead

So, practically, for the person still choosing:

Deliberate to qualify, not to rank. If your idea passes the tests, more research is just postponing the decision. If two ideas pass, flip a coin. I mean that literally — either one is fine, because the launch is what produces the answer you've been trying to research your way to.

Build the eighty percent version. Mine shipped at about eighty percent, and the missing twenty percent came from students, after launch. If you've been "almost done" for a month, that's a different problem, and I've written about it too.

And when the market answers, believe it. This is the part people skip. Launch, and you finally know something — so act like you know it, even when you don't like the answer. Feed the course that's selling, even if you'd have graded it a C. Don't fix what's working. And let the favorite go. Nobody ever bought a course because its creator was proud of it.

Still okay at best

That million-dollar course? My opinion of it never really changed. Still okay at best. My opinion just turned out not to be part of the business.

The ranking was never mine to make, and it isn't yours either. The market makes it, after you launch, in dollars. So stop trying to write it in your head. Pick a qualified idea, build the eighty percent version, and go get your answer.