Field Notes

Rent Before You Build: The Swimming Pool Problem

Tom Quinn · The AI Genie

September 4, 2026 · Updated September 4, 2026 · 4 min read

A law office once asked us to build them a custom intake app. Client fills out a form, paperwork routes itself, follow-up emails fire on schedule. Sounds like a six-month build with a developer on retainer, right?

It took an afternoon. We didn't write a line of custom code. We took three tools already sitting in their office — their scheduling app, their email service, and a form builder — and wired them together. Done by 4pm. Nobody had to learn anything new.

What does "rent before you build" actually mean?

Here's the idea, and it's not ours — it's Clayton Christensen's insight that good-enough usually beats over-built, paired with Herbert Simon's older idea of "satisficing": stop looking for perfect, take the option that solves the problem well enough, and move on with your day.

On Main Street, that means: rent the tools that already exist. Don't build your own.

Think of it like buying a car to get groceries. You don't design an engine from scratch. You buy a Honda, put gas in it, and drive to the store. Custom software is the same choice, except most owners don't realize they're being handed the blueprints for a car when all they wanted was a ride to the store.

Why is custom software a swimming pool?

A pool sounds great in June. Friends over, kids splashing, life is good. By August, you're skimming leaves. By October, you're draining it. By January, you're paying someone to keep the pump from freezing.

Custom software works the same way. It's exciting on launch day. Then it needs updates. Then the one developer who understood it leaves. Then a security patch breaks the login page and you're paying $200 an hour for someone to figure out what happened three years ago.

A rented tool doesn't have a pump. Someone else keeps it warm, patched, and running, and you just show up and swim.

What does this look like across different trades?

  • Dentist office: Instead of a custom patient portal, connect an appointment app to a text-reminder service. Patients get reminders, front desk stops making forty calls a week.
  • Florist: Instead of a custom order tracker for Valentine's week, use an off-the-shelf order form linked to a shared spreadsheet everyone on the team can see from their phone.
  • HVAC company: Instead of building a dispatch app, rent a scheduling tool that already talks to Google Maps, so techs get routed without anyone touching code.
  • Auto shop: Instead of a custom quote system, use an estimate template tool connected to text messaging, so customers approve repairs from their phone in the parking lot.
  • Hardware store: Instead of custom inventory software, use a point-of-sale system that already tracks stock and reorders for you.

None of these are exciting stories. That's the point. Boring and working beats exciting and broken.

When does building custom actually make sense?

Rarely, but not never. If you've tried three off-the-shelf tools and none of them do the one weird thing your business needs — a plumber tracking a very specific permit workflow across five counties, say — custom might earn its keep. But that's the exception, not the plan you start with.

Most owners skip straight to "we need an app" the way most homeowners skip straight to "we need a pool" before checking if the community center down the street already has one, open to the public, no maintenance required.

How do you know if you're renting or building?

SignalRentingBuilding
Setup timeHours to daysWeeks to months
Monthly cost$20–$200$0, until the invoice for "just one more feature"
Who fixes it when it breaksThe vendor's support teamYou, or whoever you can find
What happens when you growUpgrade a planRebuild a feature

A clinic we talked to had spent eight months and eleven thousand dollars on a custom scheduling tool that still didn't send text reminders. We got a rented tool doing that same job, reminders included, in under a week — for less than what they'd spent on one month of developer invoices.

A salon owner put it best: "I don't want to own the software. I want to own the salon."

What's the takeaway for a Main Street owner?

Rent the car. Don't build the engine. Skip the pool, join the club down the street. The tools already exist — most days, the win isn't inventing something new, it's picking the right three things off the shelf and bolting them together before lunch.

If you're staring down a "we need custom software" conversation right now, it's worth a second look before anyone signs a contract. We do a quick assessment, and often a $499 plan gets the same job done that a five-figure build was about to promise. No pool required.

Frequently asked

Isn't custom software better because it's built exactly for my business?

Sometimes, but "exactly right" comes with a cost most owners don't see up front: every update, every bug, every new employee who needs training is now your job forever. Off-the-shelf tools have already been tested by thousands of other businesses. You're not the first dentist office to need appointment reminders — someone already built and refined that tool. Renting means you benefit from all that refinement without paying for it.

What if the off-the-shelf tools don't talk to each other?

Most modern tools are built to connect — that's the whole trick. A scheduling app, a payment tool, and a texting service can usually be wired together with a connector tool, no code needed. That's exactly what we did for the law office: three separate tools, linked in an afternoon.

How do I know which tools to rent?

Start with the one task costing you the most time or the most missed follow-ups — late invoices, no-show appointments, slow quotes. Look for a tool built specifically for that one job, in your trade if possible. A florist's order tool and a law office's intake tool look nothing alike, and that's fine.

What happens if I outgrow the rented tools later?

You upgrade a plan or swap a tool — that's a Tuesday afternoon problem, not a six-month rebuild. Compare that to outgrowing custom software, where the fix usually means finding the original developer, or starting over.

Is there ever a case where I do need something built from scratch?

Occasionally — usually when your business has a genuinely unusual workflow that no off-the-shelf tool covers, even after trying two or three. Even then, it's worth building the smallest possible piece custom and renting everything else around it.

Want the genie to look at your three wishes?

✦ Make your three wishes