Workflow Automation Readiness Checklist (Free)

The fastest way to waste money on automation is to point it at the wrong process. Automate something that’s done rarely, follows no consistent rules, or depends on data trapped in someone’s inbox, and you’ll spend more building and babysitting the automation than you ever spent doing the work by hand.

The trick is knowing which processes are ready. This checklist scores any candidate process against 18 questions grouped into the five things that decide whether automation will stick: the shape of the process, the inputs, the rules, the numbers, and the risk. Run your shortlist through it and the answer usually sorts itself out.

How to score it

For each statement, answer Yes (2 points), Partly (1 point) or No (0 points). Be honest: the value is in the low scores, because they tell you exactly what to fix before you build. The downloaded sheet tallies the points for you.

1. The process is the right shape

It runs often and follows the same steps almost every time.
The steps are written down, or one person could write them in an hour.
It has a clear trigger that starts it and a clear point where it’s finished.
It rarely needs a judgement call that only a senior person can make.

2. The inputs and data are ready

The inputs arrive in a consistent format: same fields, same place.
The data the process needs lives in systems we can connect to.
We don’t re-key the same information between systems by hand today.
The inputs are usually correct; we’re not fixing bad data every run.

3. The rules and decisions are clear

The decisions follow rules we could explain to a new starter.
Exceptions are rare, and we know what they are.
When something is unusual, it’s safe for the process to stop and ask a person.

4. The numbers justify it

This process eats meaningful staff time every week.
Delays or errors here cost real money, customers or compliance standing.
The volume is high enough that a build pays for itself in a reasonable time.

5. The risk is manageable

A mistake here is recoverable, not irreversible.
We can check the output before it reaches a customer or the books.
Someone owns this process and actually wants it improved.
We’re automating to free people for better work, not to paper over a process that’s broken for other reasons.

What your score means

Maximum score is 36.

Score	Verdict	What to do
30–36	Ready	Strong candidate. Start here.
22–29	Nearly	Worth doing, but close the gaps first, usually “write the steps down” or “get the data accessible”.
14–21	Not yet	Tidy the process and clean the inputs before automating. Automation will amplify whatever’s there.
Below 14	Leave it	Redesign the process by hand first, or pick a different target.

The two questions that matter most

If you only look at two answers, look at these:

Are the steps written down? If nobody can describe the process in plain steps, there is nothing to automate yet: you’d be automating a guess.
Is the data accessible? The most common reason automation projects stall isn’t the logic; it’s discovering that the data lives in a system with no API, a PDF, or a spreadsheet on someone’s desktop.

Fix those two and a “not yet” process often jumps straight to “ready”.

Traps this checklist is designed to catch

Automating the exception, not the rule. If half the runs are “special cases”, you’re not automating a process. You’re trying to automate judgement, and that costs far more.
Hiding a broken process. Automation makes a bad process faster and harder to see. If the process is wrong, fix the process first.
Counting the wrong savings. “Two minutes a time” only matters at volume. Score the numbers honestly before you fall in love with the idea.

A high score doesn’t mean the build is trivial: integrations and edge cases still take real work. It means the foundations are there, which is the part you can’t shortcut.