You Owe Me for the Years, Not the Minutes. Does That Still Hold With AI?

There’s a line that’s been going around the internet for years, and I don’t know who said it first:

“If I do a job in 30 minutes, it’s because I spent 10 years learning how to do that in 30 minutes. You owe me for the years, not the minutes.”

I’ve lived by some version of that since 2020, and I’ve built my whole pricing model around it. Lately I’ve been testing it against something new: pricing your services with AI in the mix, when the tools do a big chunk of the production work and I finish things faster than I ever thought I could.

This fall I took on a few one-off hourly projects, which I don’t usually do. With AI in my workflow, I was wrapping up in a fraction of the time I would have quoted a year ago. Somewhere around the third invoice I realized the faster I worked, the less I made. I was getting my time back (the one thing I can’t make more of) and paying for it out of my own income.

So I sat with the question for a while. Does “the years, not the minutes” still hold up when AI is doing the work? Here’s where I landed.

The saying was always about results

Experience has always shown up as speed. When I started doing Dubsado setups, each one took me over 50 hours. Once I had my process documented and every task in order, that dropped to about 15. I didn’t lower my price when I got faster. I raised it, a lot, because the client was getting a better setup with fewer mistakes, built by someone who had already seen every version of their problem.

That’s what the saying protects. A client paying for a setup is paying for the system working at the end, plus the hundred small decisions that get it there. The hours were only ever a rough way of measuring that, and a bad one, since the more experienced you are, the fewer hours it takes.

What AI changed, and what it didn’t

AI changed production. First drafts of emails, formulas, documentation, the first pass at an automation: things that used to take me an afternoon now take minutes. I’ve written about how I actually use AI in my business, and the honest version is that it handles a LOT of the typing.

What it didn’t change is everything that comes before and after the typing. AI does what you tell it to do. Knowing what to tell it, what to build, what to leave alone, and which question the client should have asked but didn’t, all of that still comes from years of doing the work. And AI hands back output that looks finished whether it’s right or not. A workflow can look perfect and still send the contract before the deposit clears. Catching that takes experience, and clients can’t get experience from a subscription.

So if anything, the gap between someone with experience and someone without it got wider. Two people with the same AI tool get very different results, and the difference is judgment.

Hourly billing is where it falls apart

Here’s the problem I ran into. When you bill by the hour, every efficiency you gain is a pay cut. Learn a faster way to do something, lose money. Use a tool that saves three hours, lose three hours of income. The incentive is to go slower. My clients definitely don’t want me to be slow on purpose, and I don’t want to be the kind of person who does.

At the same time, I have to support myself. Pretending the math doesn’t matter would be naive. If a project that used to pay for ten hours now pays for three, and I still have the same bills, I either need three times the projects or a different way to price.

That’s the real answer to my own question. AI puts you at a financial disadvantage only if you’re selling hours. The hourly model was already working against experienced people, and AI just made it obvious.

My position: the saying holds, if you stop selling time

“You owe me for the years, not the minutes” is more true now than when I started believing it. What has to change is how I charge for it. Here’s what I’m doing, and what I’d tell anyone who sells their expertise.

Price the outcome and spell out what’s included

Project pricing has been my standard for years, and AI makes the case for it stronger. The client agrees to a result and a price. How long it takes me is my business. If I get faster, the margin is mine, and the client still gets exactly what they paid for. What protects you is writing down what’s included, what isn’t, and what happens when something new comes up.

If you have to bill hourly, change the terms

Some work really is open ended, and hourly makes sense for it. In that case, set a minimum (a two-hour block, for example) and raise the rate so it reflects the speed you bring. A higher rate for fewer hours is fair to everyone. Charging the old rate for a third of the time just gives your experience away.

Keep the time you save

This is the one I had to say out loud to myself. Getting time back is the whole point of building systems. It’s literally what I sell my clients. Turning around and handing that time back as a discount makes no sense. Use it for rest, for better work, or for the next client.

Be honest that you use AI

I don’t hide it. “I use AI, so you get this faster and my time goes to the work that needs a person” is a selling point. Clients are paying for the result and for my judgment on top of it. Hiding the tools makes it look like there’s something to hide.

Sell the judgment work, not the typing

Here’s the honest risk. Tasks anyone can now do in five minutes with AI are going to get cheaper, and clients will notice. If most of what you sell is production (writing the email, building the form someone else designed), that work will get repriced. The work that holds its value is mapping the process, deciding what to automate, testing it like a real client would, and handing it off so it keeps working. That’s where I spend my time, and that’s what I price.

If you’re the one hiring

If you’re a business owner bringing in help, this matters for you too. An hourly quote from someone fast and experienced can look more expensive per hour and still cost you less overall. Ask what they’re accountable for at the end, not how many hours they expect to bill. A fixed project price protects you as much as it protects them, because you know the number before you start. If you want to see how that math works for a system build, my post on calculating the ROI of a setup walks through it.

That’s how I work. Whether it’s tech stack consulting or a custom system build, you get a project with a written list of what’s included and a clear price, built by someone who uses every tool available to get it done well. The years of experience are still what you’re paying for. They just come with a faster turnaround now.

FAQ

Should I charge less if AI makes my work faster?

No. The client’s result didn’t get worse because you got faster. Price the outcome instead of the hours, and keep the time you save.

Should I tell clients I use AI?

Yes. Explain what it speeds up and what you still do yourself, like reviewing, testing and making the judgment calls. Being upfront builds more trust than hiding it.

Is hourly billing ever the right choice?

For open-ended support where you can’t list everything that’s included upfront, yes. Set a minimum block and a rate that reflects how fast you work now.

How do I switch from hourly to project pricing?

Start with your next new client. Define the result, list what’s included and what isn’t, and price it on the value of the result and what it would cost the client to keep doing it the old way.