Build vs. Buy
Build vs. buy is the ownership decision for a software capability. Framed as a binary it produces bad outcomes; the full menu is renew, renegotiate, switch, consolidate, or build. Build candidates must pass an ownership screen - stable workload, weak vendor moat, material spend, a named internal owner, no key-person dependency - and carry a five-year total cost of ownership model, because maintenance and staffing dominate the true cost of owned software.
AI-assisted development reopened this question everywhere at once by collapsing the cost of the first draft. It left the expensive parts - security, integration, monitoring, years of maintenance - unchanged. The decision discipline matters more now, not less, because the temptation arrives weekly.
Related terms
- Backsourcing — Bringing previously outsourced work - most often software development - back in-house; buyers usually say 'bring development back in-house.'
- Switching Costs — The full cost of moving off a software vendor - migration, retraining, integration rework, and risk - which determines how much negotiating leverage you really have.
- Total Cost of Ownership (TCO) — The full multi-year cost of owning a system - build, run, maintain, staff, and risk - as opposed to the purchase or build price alone.
Where this gets applied
- Unit Economics — CAC payback, NRR, gross margin by segment, cohort analysis, paid-on-bookings vs. paid-on-cash.
- Technical Debt — Quantification in dollars, not adjectives. Then a remediation plan that runs in parallel with delivery.