The last time we debated build vs buy, nothing shipped for six weeks.
I sat in a room with “Payments & Invoicing” on the whiteboard. A senior engineer said, “We can build it in a sprint.” It turned into two sprints, then three: taxes, retries, edge cases, compliance. Meanwhile, Finance needed invoices live this quarter. When we finally picked a vendor, we had burned time and trust.
We changed the approach: a one-hour decision workshop that ends with a score, an owner, and a review date.
The calm framework (score 1–5 on each)
- Differentiation: Does this make us meaningfully different?
- Time to Value: How fast can customers benefit?
- Total Cost of Ownership: People, training, maintenance, opportunity cost.
- Risk Surface: Compliance, uptime, security, vendor lock-in.
- Team Energy: Do we have the skills—and the appetite—right now?
Add the scores.
- 15 or fewer → Lean buy
- 16–20 → Mixed; pilot both or do a narrow trial
- 21+ → Strong build signal
The meeting that finally ended on time
We needed billing. Critical but not our secret sauce.
- Differentiation: 1 (customers don’t pick you for billing)
- Time to Value: 5 (buying gets you live this quarter)
- Total Cost of Ownership: 4 (vendors cost money, but maintaining your own is expensive)
- Risk Surface: 4 (vendors handle hard, regulated parts)
- Team Energy: 3 (you could build, but the roadmap is full)
Score: 17. Decision: buy—with a clear exit plan.
Decide in one hour (agenda you can copy)
- Invite one representative each from Product, Engineering, Finance, and Support.
- Write the five criteria on a whiteboard. Timebox each to 5 minutes.
- Score silently; then discuss outliers only.
- State what you’re optimizing this quarter (speed? margin? control?).
- Pick: build, buy, or pilot. Name the owner and the review date.
- Write it down in the decision log. Share it company-wide.
Guardrails that prevent regret
- Exit plan first: how you’ll leave the vendor if needed.
- Clean domain model: keep business logic yours so swapping is possible.
- Negotiated data export: format you can actually use.
- 90-day review: keep the promise—kill or scale.
When building is the right call
- The workflow is your moat and hard to copy.
- The experience is central to why customers pick you.
- Vendor constraints block essential features or pricing.
Even then, slice it. Build the minimum that delivers value now. Defer the clever parts.
A client’s quoting flow was their moat. They built two screens and a basic rules engine in four weeks, then iterated. Churn dropped; sales cycle shortened. Building the minimum let them learn fast without swallowing an entire quarter.
Share the decision in plain English
- Choice: “Buy Acme Billing”
- Why now: “Fastest path to invoices this quarter”
- Owner: “Maya (Product)”
- Review: “In 90 days; criteria = time to cash, support tickets, margin”
- Exit plan: “Export via nightly S3 dump; keep pricing rules in our system”
You don’t need the perfect choice. You need a good choice made soon—and a promise to revisit it.