How to spot a process worth automating

Not every repetitive task is worth automating. Some processes cost more to build a solution for than they'll ever save; others quietly bleed time and money with every week that passes without one. The difference almost never comes down to technology — it comes down to three simple things you can check before writing a single line of workflow.

Frequency and volume, not "how annoying it feels"

The first thing to measure isn't how tedious a task feels, but how often it actually happens. A process that runs once a month, however clunky, probably won't pay back the cost of building it. A process that runs dozens of times a day — an initial reply to a lead, a meeting booked, data moved between two systems — is exactly where every second saved gets multiplied by volume, turning into a real number within the first month.

How many rules actually decide the outcome

The second measure is how much human judgment the task genuinely requires. If the process can be described in one clear sentence — "if the lead replies yes, book a meeting; if they don't reply within two days, send a reminder" — that's a strong signal it's a great candidate. Processes that require context, negotiation, or real professional judgment need a different approach: not full automation, but a tool that speeds up the person making the decision, not one that replaces them.

What happens when something goes wrong

The third measure, and the one most businesses skip, is the cost of failure. A process where a mistake means a wrong invoice or a permanently lost lead needs to be built more carefully, with checks and stop points — but it's still worth automating, because the cost of keeping it manual is higher. A process where a mistake is a minor inconvenience fixable with one click is where you can move fast and improve as you go.

• The task repeats at least a few dozen times a week
• You can describe the rules behind it in one or two clear sentences
• You can measure the outcome (time, errors, cost) before and after
• The cost of a mistake is known and acceptable — not just embarrassing

If a process in your business checks all three boxes, that's exactly the kind of work good automation was built for. If it doesn't, that's not a dead end — it means the process itself needs simplifying first, before automation becomes the right next step.