Every team eventually faces a piece of internal software they could either build or buy. The wrong call in either direction is expensive: build something a commodity tool already does well and you have paid to maintain a worse version forever; buy something core to how you work and you have handed your differentiation to a vendor's roadmap.
The decision is usually clearer than it feels, if you ask the right questions.
Is this how you compete, or just something you need?
This is the first and most important filter. If the software encodes something specific to how your business wins — your pricing logic, your operational workflow, the thing customers actually pay you for — building it keeps you in control of it. If it is a solved problem that every company has (email, payroll, a helpdesk), buying is almost always right.
What is the real cost of buying?
The licence fee is the visible cost. The real ones are the integration work to make the tool fit, the workflows you bend to match it, and the switching cost once your data lives inside it. A cheap tool that does not quite fit can cost more than an expensive one that does.
What is the real cost of building?
People underestimate this, and it is not the first version. It is the second year — maintenance, edge cases, the person who has to own it, and the features you will keep needing that a vendor would have shipped for you. Building is not a one-time cost. It is a standing commitment.
- Buy when the problem is common, well-served, and not how you differentiate.
- Build when the software is close to your core, or when every off-the-shelf option forces a workflow that fights how you actually operate.
- Do both more often than people expect: buy the commodity pieces and build the thin layer that is genuinely yours.
The honest test
Imagine the tool exists and works perfectly, from a vendor, at a fair price. Are you relieved, or do you feel like you have lost something? Relief means buy. That flicker of "but ours would need to work differently" is usually the signal that this one is core enough to build.
Buy your commodities. Build your edge. The mistake is doing it the other way around.