In May 2002, I quit my investment banking job at JP Morgan in New York City. After years of 80-hour workweeks, I was ready for something else.

At first, I took a big detour. An uncle living in the backwoods of Mississippi had 10 acres near a gorgeous national forest, where he planned to launch a canoe rental business. Compliments of a nasty divorce, everything he owned (except for his truck) was headed for bankruptcy auction. Standing on the local courthouse steps, I bought his land, and suddenly, there I was, bootstrapping a business with my uncle.

The business was called Soggy Bottom Canoe and Kayak Rental. It was my first start-up. And it was an unmitigated disaster. Before it was all over, it would drain my investment banking savings and a good portion of my parents' savings when they tried to save the business I'd written off.

Eventually, I called it a day and moved to Silicon Valley. A marriage and four kids later, running a start-up in the backwoods of Mississippi feels like a lifetime ago. But I've realized I still apply the (expensive) lessons I learned at Soggy Bottom. Here's how you can too.

Test before you invest

The saying "if you build it, they will come" is only true in some cases. Before you invest, you've got to test whether the business you're building is what customers want.

When I arrived at the Soggy Bottom property, it was a work in progress with a half-built pavilion. For a while, I slept in a motel and then on the floor of the building. We invested heavily in finishing out our facilities and buying equipment--32 canoes, 16 kayaks, four trailers and two vans, so by the time our July 4th opening rolled around, everything was immaculate.

The "Now Open" sign went up and we waited for the crowd to rush in.

The sound of chirping crickets was deafening.

Sometimes they don't come

In our quest to have the best facilities and equipment, we neglected to speak to even a single prospective customer. No Boy Scout troops, no church youth groups, no fraternities from Southern Mississippi University. As a result, I can count on only one hand the number of times, in the seven years that we owned the business, that we reached even half of our booking capacity.

Today, I still fight against the urge to make big investments before we've "beta tested." Every time that urge pops up, I picture gleaming new canoes hitched up to sad empty passenger vans.

Fake it before you make it

Today, I know testing is critical. We describe what we are thinking of offering, gather feedback, and invite participation in a beta. We set the expectation that we have something valuable to offer, but that it's not yet well-oiled--and that their help will be a critical part of our fine-tuning process.

While start-ups may lay claim to the "beta" approach and consider it a best practice (check out this beta test page on Reddit), testing isn't a new concept. When restaurants open, they "fake open" first by holding free invite-only dinners to work out the kinkswith diners who know they're going to be guinea pigs.

At Soggy Bottom, we knew we had a wonderful stretch of water. But we didn't know that we should have spent the bare minimum effort on infrastructure. Every bit of effort remaining should have gone into letting people know about our new business, inviting them in, and hanging onto their feedback.

If we'd mocked up a full-scale operation in the short-term then built one out once we had proof of concept, perhaps we would have made it in the long run.