Every growing business hits this question eventually. The off-the-shelf tools that got you here have started to chafe, someone’s floated the idea of building something custom, and now you’re weighing it: do we keep buying software and bending to fit it, or do we build our own and own the thing outright? It’s framed as a cost question — subscriptions versus development — and that framing is where most of the bad decisions come from.
The question most people actually ask
The usual version is “which tool should we buy?” — a comparison of features and prices across three products. It’s the wrong starting point, because it skips the only question that matters first: should this part of your business run on a tool at all, or is it specific enough that it deserves to be built? Get that wrong and the best tool in the category still won’t fit, because the category itself was the wrong shape.
So before comparing anything, separate your operation into two kinds of work: the parts that are the same in every business like yours, and the parts that are yours alone.
Where buying wins
For commodity functions, buying is almost always right, and building is almost always a mistake. Email delivery, payment processing, accounting, calendars, video calls, file storage — these are solved problems, maintained by companies that do nothing else, at a price you could never match by building. There’s no advantage hiding in your version of payment processing. Buy these, and don’t let anyone talk you into reinventing them.
The test is simple: if a function is identical across your whole industry and nobody chooses you because of how you do it, buy it.
Where building wins
Building earns its cost in exactly the opposite place: the parts of your operation that are specific to how you work, and especially the parts that are the reason customers choose you. The particular way you qualify and route a lead. The workflow that is, quietly, your actual business. The handoff between two teams that off-the-shelf tools force you to do by hand. And — this one is underrated — the glue between your bought tools, so they behave as one system instead of five islands.
The test, again, is simple: if a function is core to how you operate or how you compete, and no tool fits it without forcing you to change how you work, that’s a candidate to build.
The hidden cost of “just buy it”
Buying looks cheaper because its cost is a number on an invoice. The real cost is everything you do around the parts that don’t fit: the manual re-typing between apps, the workarounds your team maintains, the process you reshaped to suit the software, the lock-in that makes leaving expensive, and the quiet ceiling the tool puts on how you can grow. None of that shows up in the monthly price. All of it compounds.
The cost of buying isn’t the subscription. It’s the price of bending your operation to fit it — paid every day, forever.
Building has the opposite cost shape: a larger, visible cost up front, and a smaller, shrinking cost after — no per-seat creep, no workaround tax, no rebuild when you outgrow someone else’s roadmap. Whether that trade is worth it depends entirely on how core the function is, which is why the build-vs-buy question can’t be answered by price alone.
It’s rarely all-or-nothing
The honest answer, for almost every business, isn’t “buy everything” or “build everything.” It’s a fitted system: buy the commodity pieces, build the parts that are genuinely yours, and connect them so the whole thing runs as one operation. The skill isn’t in choosing a side — it’s in drawing the line in the right place, function by function, and then engineering the seams between bought and built so a customer or your team never feels the join.
That’s also what separates a system from a pile of subscriptions. Most businesses already own plenty of good tools. What they’re missing is the architecture that makes those tools, plus the few custom pieces that matter, behave as a single coherent operation.
How to actually decide
For any function you’re unsure about, work through four questions:
- Is this core to how we compete, or is it plumbing? Plumbing gets bought. Differentiators are candidates to build.
- Does any existing tool fit without forcing us to change how we work? If yes, buy it. If every option requires bending your process, that’s a signal the fit isn’t there.
- What’s the real three-year cost of buying? Not the subscription — the subscription plus the workarounds, the manual labour, and the ceiling it imposes.
- What happens when we grow? Does this choice scale with you, or does it become the thing you have to tear out in two years?
Answer those honestly across your operation and the line between buy and build tends to draw itself. The mistake is treating it as a single decision for the whole business, when it’s really a series of small, deliberate calls — and then a matter of engineering the pieces into one system.