Most founders ask the wrong build-vs-buy question. They ask “which is cheaper?” The right question is “which one compounds?”
Commodity software gets cheaper the longer you rent it. A workflow that encodes how your team wins gets more valuable the longer you own it. We learned this building our own CRM instead of buying one: the 20-minute manual pipeline review became a 30-second AI-assisted review, data entry dropped 95%, and each rep got back 20+ hours a week. No off-the-shelf CRM would have done that, because no off-the-shelf CRM was designed around our sales motion.
Short answer: buy commodity, build edge. Everything below is how to tell which is which.
The decision in one table
| Factor | Buy SaaS | Build it |
|---|---|---|
| Problem is standard (auth, billing, email) | ✓ | |
| Workflow is your competitive edge | ✓ | |
| AI leverage on your own data | ✓ | |
| Off-the-shelf needs constant workarounds | ✓ | |
| 3-year SaaS cost > one focused build | ✓ | |
| You need it working this week | ✓ | |
| No engineer to own it afterward | ✓ |
If three or more rows point to build, scope it. If not, buy and move on.
Why SaaS workarounds are the hidden tax
The sticker price of SaaS is honest. The workaround cost isn’t. Every “we’ll just export to CSV and…” is engineering time, stale data, and reps who stop trusting the system. Before comparing prices, audit the workarounds: list every manual step your team performs around the tool. If that list costs more than 5 hours a week, the buy option is already more expensive than it looks.
This was our CRM math. The SaaS options were fine products — they just assumed a generic sales process. Ours isn’t generic: AI enrichment, intent scoring, and drafted outreach are the workflow. Bending a generic CRM into that shape would have cost more than building.
What to build: one workflow, not a platform
The failure mode of internal builds is building a platform. The fix is a written scope contract:
- One must-win workflow. Ours: pipeline review. Everything serves it.
- Explicit non-goals. Reporting dashboards, mobile app, integrations we don’t use — written down, not built.
- Fixed slices. UI, API, data, AI — each priced and shippable independently.
- Boring stack. Next.js, Postgres, Tailwind. Speed of change beats architectural fashion for internal tools.
Our CRM loads in under 50ms because there’s nothing to be clever about. Internal tools should be fast and easy to change — that’s the whole spec.
What it costs (honestly)
A focused internal tool is a 2–4 week build, not a quarter-long project. Compare that one-time cost against 36 months of per-seat fees plus workaround labor. The breakeven is usually 12–18 months. After that, you own an asset instead of renting a constraint.
What building costs that nobody budgets: ownership. Someone has to keep it running. Budget half a day a month, keep the stack boring, and this stays trivial. Skip ownership and even a good tool rots.
When we would not build
Don’t build when the problem is commoditized (billing, auth, transactional email), when you need it this week, or when nobody will own it afterward. Don’t build two internal tools at once — finish one, measure the payoff, then decide.
Decision checklist
- The workflow differs meaningfully from the generic SaaS assumption
- Workarounds cost 5+ hours/week today
- 3-year SaaS cost exceeds a 2–4 week focused build
- One must-win workflow is written down, with non-goals
- An owner exists for maintenance (half a day/month)
- The stack is boring enough to change quickly
Tick five of six? Build it. Otherwise, buy — and spend the saved months on the product customers pay for.
What this means for founders
Internal tools are leverage or distraction, nothing in between. One sharp tool that removes a bottleneck (like our CRM’s 30-second review) pays for itself in weeks. A sprawling internal platform nobody asked for is a second startup you’re funding accidentally. The discipline isn’t engineering — it’s saying no to everything except the one workflow that matters.
Related: the 10-Day MVP Scoping Playbook is how we scope builds like this, the CRM case study shows the full before/after, and the MVP Scope Worksheet is the template we start from.