How Do I Pick the First Process to Automate?

How do I pick the first process to automate?

Straight answer

Pick a process that is frequent, repetitive, rule-based and forgiving of mistakes. List your candidates, score each on how often it runs, how well-defined it is, and how costly an error would be, then choose the one with high frequency, clear rules and low risk. That combination gives the fastest, safest first win.

Information current as at 5 July 2026

Automating the wrong process first is how good automation projects earn a bad name. The right first choice is not the biggest or most impressive; it is the one that gives a clear win with little risk. There is a simple shape to look for, and a simple way to score your candidates against it.

Plain English

Process: A repeatable sequence of steps your business follows to get something done.
Rule-based: A task that follows clear, consistent rules rather than needing case-by-case judgement.
Exception: A case that falls outside the normal rules and needs a human to handle it.
Volume: How often a process runs, which decides how much time automating it saves.

Step by step

List the processes that eat your timeWrite down the repetitive processes across your business: the routine emails, the data moved from one place to another, the reports compiled the same way each week, the requests handled with the same steps every time. Do not filter yet; just get them on a list. You are looking for the recurring, patterned work that quietly consumes hours. The processes worth automating are almost always the dull, frequent ones, not the interesting, occasional ones.
Score each on frequencyFor each candidate, note how often it runs: many times a day, weekly, monthly. Frequency matters most, because a process you do fifty times a day saves real time when automated, while one you do twice a year barely moves the needle even if it is annoying. High-frequency processes are where automation pays back fastest, so let this be the first thing that pushes a candidate up or down your list.
Score each on how well-defined it isNext, judge how rule-based each process is. Does it follow clear, consistent steps every time, or does it need judgement and vary case by case? Clear rules automate cleanly; heavy judgement does not. Also note how many exceptions crop up, the odd cases that break the pattern, because a process riddled with exceptions is far harder to automate reliably. Favour the ones that follow the same predictable path almost every time.
Score each on the cost of a mistakeNow weigh the risk. If the automation got something wrong, how bad would it be? A wrongly sorted internal file is trivial; a wrong invoice or a mishandled customer is not. For your first automation, deliberately choose something forgiving, where an error is caught easily and costs little. Save the high-stakes processes for when you have experience and trust in how automation behaves. Low risk first is not timidity; it is how you avoid a first failure that sours everyone on the idea.
Choose the high-frequency, well-defined, low-risk oneNow look at your scored list and pick the process that is run often, follows clear rules, and would not hurt much if it erred. That intersection is your first automation. It may not be the most exciting candidate, but it will give you a fast, visible, safe win that builds confidence and evidence for the next one. Automate it, keep the old way running alongside at first, and measure the time it hands back before you move on.

No pressure

Show us what you built.

If you have made something and it needs to become real, send it over. We will tell you honestly what it needs to be live, safe and yours, whether that is a quick fix you can do or a proper build. No obligation.

Questions, answered

What makes a process a good candidate for automation?

Four things together: it runs frequently, it follows clear rules, it has few exceptions, and a mistake would be cheap to fix. A process with all four gives a fast, safe win. Missing any one makes it a poorer first choice, especially high risk or heavy judgement, which are the two that most often turn automation into trouble.

Should I automate my biggest, most painful process first?

Usually not first. The biggest processes tend to be complex, high-stakes and full of exceptions, which makes them risky as a first attempt. Start with something smaller, clearer and more forgiving to build confidence and evidence. Once you understand how automation behaves in your business, the bigger process becomes a sensible next target rather than a risky opening move.

What if a process has lots of exceptions?

Then it is harder to automate reliably, because each exception is a case the automation may get wrong. You can sometimes automate the common path and route exceptions to a human, but a process that is mostly exceptions is a poor first choice. Prefer one that follows the same predictable steps nearly every time for your opening automation.

How do I measure whether the automation was worth it?

Decide the outcome up front, usually time saved or errors reduced, then compare before and after over a few weeks. Count the hours the process used to take and what it takes now, including any checking. A clear time saving on a frequent process is the sign it was worth it. If the numbers barely move, choose a different candidate next.

No pressure

Show us what you built.

How do I pick the first process to automate?

Step by step

Questions, answered

Two doors. Same senior team.