Build vs. Buy: buy the boring stuff, build what makes you different

Here is the whole decision in one line, and saying it out loud costs us work: buy the commodity, build only what makes you different. Everything else is noise.

Think about what your business actually runs on. Most of it is not special, and it was never supposed to be. Email, calendars, payroll, accounting, file storage, video calls. Everybody does these the same way, the tools already exist, and they are cheap. Building your own version of any of them is a great way to set money on fire and lose a few months doing it. If a task is common, and every business on earth does it more or less the same, buy the tool and get on with your day. We will tell you that to your face, even though we are a custom shop and it means less work for us.

The question only gets interesting for the handful of processes that are actually yours. The way you quote a job. The checklist that keeps your mistakes out. The workflow you spent years figuring out that is the real reason a customer picks you over the shop down the road. That is where "just buy it" quietly starts costing you.

Off-the-shelf software is built for a market, not for you. A vendor has to sell the same product to ten thousand companies, so it gets shaped around what those ten thousand have in common, not the thing that makes you you. It will do eighty percent of what you need, and do it well.

Then you hit the last twenty percent, the part that is actually your edge, and the wall goes up. You ask for it, and now you are one ticket in a line ten thousand deep, waiting on a roadmap that may never reach your name.

So you make do. You bend your process to fit the tool, or you pay consultants to force the fit, or somebody on your team spends every Friday afternoon copying numbers between two systems that will not talk, and has for years.

None of that is on the sticker price. You pay it anyway, every month, in someone's time and patience.

Building is not free either, and anyone who tells you it is has their hand in your pocket. There are two costs: building the tool, and keeping it running. How big that second one gets depends entirely on what you build. A sprawling, do-everything system is a bill that never stops coming. A small tool that does one job can run for years with barely a touch.

That is the part off-the-shelf can't give you: control over which of those you get. A tool scoped to one job rarely changes. It is not a platform with a whole team behind it, so months can go by without anyone touching it.

You decide how simple it stays and what it is responsible for. Done right, the gap your team patches by hand shrinks to almost nothing, instead of the permanent gap the shelf tool leaves you filling every week.

Go in with your eyes open, someone still owns the upkeep, and you want it built simple enough that one person leaving doesn't take the whole thing with them. But "custom means expensive forever" is a myth. Custom means you decide what it costs to keep alive.

So the real answer is not build or buy. It is both. Buy the foundation, build the part that is yours. Nobody builds a basic calculator to run simple arithmetic. They pull up the app on their phone and move on.

Same idea, scaled to your business: use what already exists for the ordinary stuff, and spend your time and money only on the build that fits your process. Then wire the two together.

Now, when I say rent the boring, do not hear it as "so the part I build should be the exciting one." It almost never is.

The builds that give you the most back, and keep giving it back, are the simple ones.

Simple is a different animal from boring. Boring is what ten thousand companies share. Simple is small, specific, and shaped like your business, which is exactly why no vendor ever packaged it. It is not a commodity, and it is not glamorous. It is usually the best thing you will ever build.

Run the numbers on that, because this is where it gets interesting. A week every month is twelve weeks a year, a full quarter of somebody's working life going into one task. About two hours a month is roughly three days a year.

Now put a salary on it. If that person earns eighty thousand dollars, a quarter of their year is about eighteen thousand dollars, and more than that once you count what they actually cost you in taxes and benefits. Be honest about what that number is, though: it is not cash coming back, you are paying the salary either way. What you get back is twelve weeks of a person you already employ. So price a build against it. At thirty thousand, it has paid for itself inside two years. At forty thousand, a little past two. After that it keeps handing the twelve weeks back every year, against upkeep that on a build this simple rounds to nothing.

And the hours were only part of the bill, because the sheet all that typing produced was never right in the first place. That is the whole formula, and it runs on your business the same way: what an automation hands back to you, multiplied by how often you do it. Your simple jobs run constantly. That is exactly why you stopped noticing them.

There is one more line on this bill, and nobody ever writes it down. I have watched someone lose sleep over a set of numbers that would not reconcile, awake in the middle of the night over a report that was never going to add up, because the process itself was broken and no amount of care was going to fix it. Nobody is proud of the week they spend retyping figures that still do not match at the end of it. That work follows people home. They get short with each other over it, and the good ones eventually leave over a job that is mostly this. None of it appears in a budget. You are paying it anyway.

Then it keeps going, and that is the part you should care about most. Simple work is stable work. The way that task got done was not going to change much the following year, so the tool built for it did not need to change either. No roadmap, no version two, nobody to retrain, nothing to relaunch. Nobody fights you on one of these either, because you are not asking your team to work a new way, you are taking away the part of their month they already hated. In year three it is still handing back those twelve weeks, and everyone has long since forgotten it is there.

That is what sustained value actually looks like, and it looks like almost nothing.

Now, none of this makes the big, ambitious system a bad idea. Sometimes the big one is exactly right, and exactly what you need. But the way you get there is usually to clear your simple wins first. The most impressive system in your industry does you no good if your people are still the bottleneck feeding it by hand. Take the certain wins, let them stack, and go after the ambitious build from a position where your team actually has the hours to adopt it.

When a client is genuinely torn, we boil it down to six questions:

1. 1
   Is this a common task, or is it how you compete? Common, lean buy. Core to how you win, lean build.
2. 2
   How often does it run? Once a quarter, the payoff is small no matter how much it annoys you. Every week, the arithmetic changes completely. Frequency is what turns a small, unremarkable job into the best build on your list.
3. 3
   Would you be OK with the vendor selling this exact thing to your competitor tomorrow? If you are, it is a commodity, buy it. If that stings, it is your edge, build it.
4. 4
   What is the five-year cost, not the day-one cost? Quick math: the monthly fee times sixty, plus what the gap costs you in workarounds and staff time. And that fee is not fixed, it usually climbs the moment your first contract term is up. Put the real number next to build-plus-upkeep, and it is rarely as close as the sticker makes it look.
5. 5
   If you build it, can you live with maintaining it, or paying someone to? If that is a no, buy it, or find a builder who will keep it running for you.
6. 6
   If the vendor never ships the one feature you actually need, can you live with that? If not, do not bet your business on their roadmap.

We are a custom shop, so you would expect us to say build, build, build. We don't. We tell people to buy the boring stuff every business shares. Then we help them pick the right one, and we build only the piece that is genuinely worth owning, which is usually the least impressive thing on the list. When we do build, we start with a small working version of your real problem before you have paid us a dime, so you are buying something you can watch work, not a promise. And whatever we make is yours alone. It never turns into a feature we sell to the shop across the street next quarter.

So if you are staring down a buy-or-build call right now, tell us the process you are weighing. You will get a straight answer on which way we would go, even if that answer is that you do not need us for this one.

Written by Stewart Liberti-Wilson