The Shed I Built in a COVID Summer

The summer of 2020 left me with a lot of time. I used it to build a 12x16 shed, by hand and to code.

I didn't start with lumber. I started with research. My town has its own rules on placement, size, setbacks from the property line, even material choice, and none of it is negotiable. So I read the ordinance, then designed the whole shed in SketchUp before I bought a single board: framing, roof trusses, door and window openings, flooring, foundation supports, all of it modeled and dimensioned.

SketchUp model of the shed's skeleton frame, showing wall studs, roof trusses, and door and window openings — The full SketchUp model, framed and dimensioned before a single board was cut.

That model became the permit application. I submitted detailed drawings, dimensions, and truss design to the town, and got approved on the first pass.

From there it was more research: how to frame a shed properly, which fasteners, which materials, aimed past code rather than at it. I built pricing spreadsheets to track every purchase and compare suppliers. I spent the money on the framing and the fasteners and saved it on the finish.

The build itself took about a month and a half, mostly solo, with my wife and daughters pitching in when I needed extra hands. By the end I'd found the right tool for nearly every job.

The finished 12x16 shed, painted and complete, sitting in the backyard — Finished, six weeks later.

Nothing went exactly to plan. The windows I'd designed around weren't in stock when I went to buy them, so I adjusted on the fly and ended up with different sizes. I hadn't accounted for how genuinely awkward it is to roof a building alone. And the biggest variable of all was weather: rain delayed me, heat delayed me, and I just had to absorb it and keep moving.

It's how I approach any big project I haven't done before, shed or software system. Read everything first, then build past the minimum, then adjust when the plan turns out to be wrong somewhere.

What I keep coming back to is how visible the progress was. On a software project, you can end a long day and genuinely wonder what you accomplished. Framing a wall doesn't have that problem. At the end of the day the wall is standing there.

In the end, it cost a fraction of what a comparable delivered shed would have run me, and I built something that will outlast the receipt.