How to choose between platform and build

Should we buy a CI platform or build it ourselves on Jira / SharePoint / spreadsheets? A practitioner's framework for the decision.

The build temptation

Every CI lead has been here. We've got Jira. We've got SharePoint. We've got a Power BI dev. Surely we can stitch this together ourselves.

Sometimes the answer is yes. Most often the answer is a slow no — three years in, you've built something that works for one team, breaks at scale, and depends on three people who can't take leave.

The four questions

How many sites or teams? Below 50 users, build can work. Above that, the integration overhead grows superlinearly.

Who owns it? If the owner is a CI lead with no engineering reports, build will struggle. CI tooling needs maintenance, not just construction.

What audit requirements? If you'll be asked for end-to-end audit trail by ISO, finance, or regulator, build it from day one — retrofitting audit is brutal.

What's the half-life of the team? If the people who build it move on within 2-3 years, the platform becomes a museum piece.

When build does work

Single team, narrow scope, in-house engineering capacity, low audit requirements. That's a reasonable build case.

Multi-site, multi-function, governance requirements, no in-house engineering capacity? Buy. The platform pays back inside year one if the alternative is a stalled CI portfolio.