Twenty Years of Building. What I'd Tell My Younger Self.

Tiffany and I started Kasteel in 2004 with a pickup truck, a borrowed table saw, and one client who took a chance on us. Twenty years on, the truck has been replaced four times, the table saw is on its third life, and the client is still a client. That last fact is the one I think about the most.

What I'd tell the 2004 version of myself

Most of what I'd say is operational. Set up your books from day one a cleanly accounted year-one looks the same as year twenty. Keep your subs paid weekly even when the client is paying you monthly; you'll learn loyalty isn't bought, but it does need to be respected. Pay yourself a salary so you can see whether the business is actually profitable, rather than mistaking a healthy bank balance for one.

But the part I'd really make sit down and listen is about clients. I'd tell him: you're going to want to take every job in the first three years. Don't. The wrong client will eat you whole emotionally, financially, reputationally and at year five you'll wish you'd said no. The fit conversation is the most important conversation in any build, more than the bid, more than the design. Have the courage to walk away early.

What's changed in twenty years

The clients have changed. In 2004, most of our work was for first-generation Atlanta money buying a second home in North Georgia. In 2024, more of the work is for second- and third-generation clients, often referred from the first set, building homes that are explicit legacy projects the place the grandkids learn to swim, the dining room that seats fourteen for Thanksgiving forever. The conversations are different. The horizon is longer.

The materials have changed. Cedar got worse. Stone got more expensive. Steel and lumber went on a rollercoaster I don't think we've fully gotten off of. Standing seam went from "a niche choice" to default. Smart-home wiring went from "oh, are we doing that" to "of course, here are the seven systems we typically integrate." The houses have gotten more complex by every measurable axis except square footage.

The crew has changed in the best way. We have a project manager who has been with us seventeen years; a superintendent who started as a framer in 2009 and now runs sites; a finish carpenter who came on for a one-month stretch in 2007 and is somehow still here. The same people show up every Monday, in the same trucks, doing the same kind of work for clients who now know them by name. That continuity is the actual product.

What hasn't changed

We still build one project at a time per project manager, capped at six concurrent residential builds and three concurrent commercial. We still do itemized bids. We still send the Wednesday photo report and we still make the Wednesday phone call. We still pay our subs weekly. Tiffany still runs the books and still catches every error before I do.

We still don't take work that doesn't fit. Some years we say no to more projects than we say yes to. The math sounds stupid until you watch what happens to the firms that say yes to everything: they grow fast for two years, hit a cash crunch in year three, and either get acquired or quietly fold. The disciplined growth has been better for our crew, our clients, and our family.

The hardest year

2009. A client who had signed in 2007 and started construction in 2008 lost the financing for the back half of the build mid-project. We had a substantially-framed home, a six-figure stop-work, and a young business with no buffer. We worked it out the client found alternative financing, we extended terms on our end, and the home was finished in 2010 but the lessons changed how we operate. Reserve cash. Monthly draws structured against actual completion percentages, not anniversary dates. Conditional clauses for major financial events. We've never had another client default since, but the playbook is in place if it happens.

The thing I didn't expect

How much of this work is, in the end, about trust. Not about wood or stone or windows or systems those are the medium. The work is keeping a promise to a family across eighteen months, in conditions that change weekly, when material prices move, when a sub no-shows, when a permit takes six weeks longer than it should. The work is, every Wednesday, telling the truth and showing up.

If you do that for twenty years, the clients you started with are still your clients, and they bring you their kids. We're starting to build for those kids now. That is the metric I would have not understood at 28 and barely understand at 50: the success of a custom builder is whether the family is still calling you in fifteen years. Everything else is implementation detail.

Here's to the next twenty.