The default should be buy
Let’s start honestly: most software problems are solved problems. CRM, accounting, HR, ticketing — buy them. Anyone who tells you to build a bespoke CRM is selling you a bespoke CRM.
Operational software is where the calculus genuinely changes.
Why operations break the rule
The tools that run heavy operations — monitoring, dispatch, asset management, planning — sit directly on top of how your operation works: your assets, your telemetry, your constraints, your failure modes. Generic products handle the generic 80%. The problem is that operational value concentrates in the specific 20% — and that’s exactly the part vendors can’t know in advance.
The result is familiar to most operations teams: an expensive platform, a spreadsheet ecosystem growing around it to cover the gaps, and operators keying data between the two.
A practical test
We use three questions:
- Is the workflow a differentiator or a commodity? If your competitors do it identically, buy. If your edge lives in how you route, schedule, monitor, or respond — that’s a candidate to build.
- Where does the data live? If the tool must ingest your live telemetry and speak to your legacy systems, integration is most of the work anyway. Owning that layer usually pays.
- What does the workaround cost? Count the spreadsheets, the re-keying, and the decisions made late because data arrived late. That, annualised, is your build budget.
The middle path
Building doesn’t mean building everything. The pattern that works is a thin, custom operational layer — the dashboards, integrations, and decision tools specific to you — composed over solid off-the-shelf infrastructure: managed databases, cloud platforms, proven open source.
That keeps the custom surface small, the maintenance honest, and the value concentrated where it belongs: on the decisions your operation makes every hour.