Published · 20 January 2026
Where AI actually pays for itself in a small business
Most AI projects fail on the wrong question. Here is the test we use before recommending anything.
The question is not "what can AI do?" — it is "which work in this business is repetitive, high volume and low judgement?" Those three conditions are where automation pays for itself, and almost nowhere else.
In practice that means routine customer enquiries, first-response on leads, recurring reporting, data entry between systems, and document drafting from templates. Work that changes every time, or that depends on a relationship, is usually a bad candidate — no matter how tempting the demo looks.
A simple test: count how many times the task happens per month, multiply by the minutes it takes, and compare that against the cost of building and running the automation. If the payback is longer than a year, it is not the place to start.