Build, buy, or automate: the right call in the age of AI

There used to be two answers when a business hit the limits of a spreadsheet: build custom software, or buy a SaaS tool. In 2026 there's a third, and it's reshaping the decision — automate it with AI. Knowing which of the three fits a given problem is now one of the most valuable judgment calls a company can make.

Buy when the problem is common

If thousands of companies have your exact problem — invoicing, email, scheduling, CRM — someone has already built a better tool than you will, and they maintain it for a living. Buying is almost always right here. Custom-building a worse version of an existing product is the most expensive mistake we see.

Build when the problem is yours

Custom software earns its cost when the problem is specific to how you work and that difference is a genuine advantage. A workflow no tool models, a product that is the business, an integration nobody sells. Here, off-the-shelf forces you to bend your business around the software instead of the other way around.

• ·The process is a competitive edge, not a commodity
• ·You need to own the data, the logic, and the roadmap
• ·Existing tools force ugly workarounds your team quietly hates

Automate when the work is repetitive and rule-ish

This is the new category. A lot of work isn't a product at all — it's a person copying data between tabs, triaging an inbox, summarising documents, answering the same question fifty times a day. AI agents and automations now handle a large share of that, often in days rather than months, and often without building a full application.

The question is no longer just "build or buy?" It's "does a human even need to be in this loop?"

How we decide with clients

We start from the problem, not the technology. We ask how common it is, how much of it is judgment versus repetition, and what owning it is actually worth. Often the answer is a mix: buy the commodity parts, automate the repetitive middle, and build only the thin slice that's genuinely yours. That's usually the cheapest path to the best result — and it's the opposite of selling you the biggest project.