Build vs buy: should you build custom software or buy off-the-shelf?

Most "should we build or buy?" debates go in circles because they treat it as one decision. It isn't. It's a decision you make per system — and the right answer is usually a mix.

Here's the framework we use with clients.

Buy when the problem is generic

If thousands of companies have the exact same need, someone has already built a mature product for it. Email, accounting, payroll, calendars, video calls — buy these. You will never out-build a focused product company on a commodity problem, and you shouldn't try.

Signs you should buy:

• The process is the same as it is for everyone else.
• A well-reviewed tool already covers 90% of what you need.
• The work isn't a source of competitive advantage — it just has to function.

Build when the workflow is your edge

The opposite is true for the handful of processes that are your business. A real estate agency's listing-to-deal pipeline. An events company's booking and logistics flow. A studio's member and class system. These are where you compete, and they're exactly where generic tools force you to compromise.

If a workflow is core to how you make money, owning it is owning your advantage.

Signs you should build:

• You're stitching three or four tools together with manual exports and copy-paste.
• Your team has "the way we actually do it" that no tool quite supports.
• You're paying per-seat for software you've outgrown, or that you have to fight every day.

The cost nobody puts on the invoice

The honest comparison isn't subscription vs build cost. Off-the-shelf has a hidden bill: the work your team does to fit the tool instead of the tool fitting them. Re-keying data between systems. Workarounds that only one person understands. Reports you export and rebuild in a spreadsheet every week.

That cost compounds quietly, and it's the reason "cheap" software often turns out expensive.

The middle path most companies should take

You don't have to choose globally. The pragmatic answer for almost everyone is:

1. Buy the commodities. Don't build your own email or accounting.
2. Build the one or two systems that are your edge. Own those completely.
3. Connect them. A good custom system pulls the bought tools into one pipeline that reports back to you.

That's the approach we take at MUBE: we don't replace everything, we build the core that's actually yours and integrate the rest around it — on a stack you own outright.

How to decide this week

Pick your most painful process and ask three questions: Is this generic or specific to us? Is a mature tool already doing it well? Are we paying for it in manual work? If it's specific, underserved, and costing you in hidden labour — that's your first candidate to build.