Prove it small, then go big

Everything in this series so far has been about starting small: the 3am list, the time-tax, the relief-to-effort ratio. Here’s the part that surprises people: starting small is also the fastest route to the big stuff.

The AI proof of concept buys the real currency: trust

Early in my career I watched a bank merger get projected to require ten times the reporting staff. We absorbed the entire load with the team we had — but only because years of small, proven automation had already built the foundation and, more importantly, the organization’s trust in it. Nobody hands the big job to systems that haven’t earned it. Nobody should.

There’s decades of research behind this instinct. Harvard Business School’s Teresa Amabile analyzed nearly 12,000 workday diaries and found that nothing drives motivation like visible progress in meaningful work — the “power of small wins.” It’s as true for an owner learning to trust automation as it is for the team living with it: each small, visible win makes the next, bigger step feel natural instead of reckless.

The same rule holds at small-business scale. The first watchdog that catches a real conflict changes how an owner sees automation. The first month the books reconcile themselves, the skepticism drops another notch. Each shipped win is a deposit in an account you’ll draw on later — when the project is bigger, scarier, and touches the parts of the business that really matter.

The foundation is technical, too

It’s not just psychology. The small projects literally build the infrastructure the big ones need. The calendar integration behind a conflict watchdog is the same plumbing a smart-scheduling system runs on. The payment matching that cleaned up the books is the data layer for real financial forecasting. In my own companies, the platform that now handles five times the original client volume wasn’t designed in one heroic stroke — it accreted, one proven block at a time, each block paying rent from the day it shipped.

Do it in the other order — grand architecture first — and every assumption is untested, every integration is new, and the whole thing has to work before any of it pays. That’s not ambition. That’s a lottery ticket with a consulting invoice attached.

When the big ideas get their turn

And they do get their turn. Once the day-to-day runs itself — once the owner isn’t firefighting, once the data is clean because the systems keep it clean — the ambitious conversation changes completely. You’re no longer asking “could AI transform this business?” over a blank whiteboard. You’re asking “given everything that already works, what’s the next block?” The answers get bold precisely because the foundation is boring.

That’s the whole philosophy. Resolve the pain. Remove the friction. Prove it, earn it, then climb. The most improvement to the owner’s life for the least effort — repeated until the grand vision isn’t a pitch anymore, it’s just the next step on a ladder you’ve already built.

This closes the series. If you missed the beginning, start with the thesis — or skip the reading and bring me your 3am list.