The answer is almost never as obvious as the SaaS sales page makes it seem.
I have had this conversation dozens of times. A company is doing something manually that could be automated. They ask should we build it ourselves or buy a tool. The question sounds simple. It is not simple.
The buy case is easy to make. Someone sells a tool that does eighty percent of what you need. It costs a hundred dollars a month. It works on day one. The sales page shows a beautiful dashboard. The demo is smooth. It is very tempting.
The build case is harder to make but often more honest. Building takes longer. It requires someone who understands both the workflow and the technology. But it does exactly what you need. Not eighty percent. One hundred percent.
The question I ask first
Before build vs. buy, I ask a different question. Is this workflow core to your business or adjacent to it?
If it is adjacent — if it is something every business in your industry does the same way — buy. Do not spend engineering time on invoice processing or standard CRM workflows. Someone has solved this. Pay them and move on.
If it is core — if it is the thing that makes your business different from competitors — build. Because the tool will never do exactly what you need. It was built for the average case. You are not the average case. You are you.
This is where most companies get it wrong. They buy tools for core workflows because buying is faster. Then they spend months customizing the tool, writing workarounds, and training people to use it in ways the vendor never intended. The total cost ends up higher than building. And the result fits worse.
The hidden cost of buying
When you buy a tool, you inherit someone else’s assumptions about how your workflow should work. Sometimes those assumptions match yours. Often they do not. And changing them is expensive — either in customization fees or in the slow accumulation of workarounds.
I have seen companies spend forty thousand dollars a year on a tool that requires ten hours a week of manual workarounds. The tool does eighty percent of the job. The remaining twenty percent falls on a human who has to bridge the gap between what the tool does and what the company needs. That is not automation. That is outsourced manual labor.
Building has the opposite problem. It takes longer upfront. It requires someone who can write the automation and maintain it. But when it is done, it fits exactly. And when the workflow changes, you can change the automation without waiting for a vendor roadmap.
The practical answer
In practice, most companies should buy for standard workflows and build for custom ones. But they should be honest about which is which.
Lead scoring is usually adjacent. Buy something that does it well enough. Your unique client onboarding sequence that involves three internal systems and a manual quality check? That is core. Build it.
The mistake is buying for core workflows because buying feels like progress. It feels like you solved the problem. You did not. You deferred the problem to a monthly subscription and a set of workarounds.
Buy for the workflows that make you normal. Build for the workflows that make you better than normal.