Build vs. buy: when to use Make / Zapier and when to build custom
A decision framework for $1M–$50M operators trying to figure out if they should buy a $99/mo SaaS or build something properly.
There's a place for Zapier. There's a place for Make. There's a place for a $99/mo SaaS. There's also a place for actually building something that fits your business. Knowing which is which saves a lot of money.
Buy when…
- It's a generic problem 10,000 other businesses also have (basic email automation, calendar reminders, simple form-to-CRM).
- The flow is shallow — under 5 steps, no branching, no judgement calls.
- You're early and just need to validate the workflow exists before investing in it.
Build when…
- The flow has branching logic — 'if this kind of customer, do X, otherwise Y' — that off-the-shelf tools can't capture cleanly.
- It touches 3+ systems and the data has to stay consistent across all of them.
- It involves judgement (writing in your tone, classifying messy customer messages, prioritising work).
- Volume is high enough that the per-action cost of a SaaS exceeds the cost of running it yourself.
The Zapier trap
Zapier is great until it isn't. The trap is the moment you have 14 zaps that no one understands, half of them silently broken, and the new hire has no idea how anything connects. We unwind a lot of these. It's almost always cheaper to build properly than to keep extending the duct tape.
Zapier is for validating. Build is for running. Don't run on Zapier past 12 months.
What we recommend at each stage
Under $1M: buy. Use SaaS. You don't have the volume or complexity to justify custom. $1M–$5M: hybrid. Custom for the 1–2 things that move the number. SaaS for the rest. $5M+: build the systems that drive your business; buy the rest. Don't buy a SaaS for your competitive moat.