Which Tasks to Automate First: A Guide for Ops Managers
November 21, 2026
To decide which tasks to automate first, list every repetitive task, record the weekly hours each consumes, score how easy each is to automate from one to five, and multiply the two. Start with the highest score, check it for risk, and run it for 30 days before moving on.
[Owner note: the brief asks for an audience named in config.py, which I could not see. This draft is written for operations managers at small and mid-sized businesses. Swap in the audience from config.py and adjust the examples if it differs.]
The question operations managers actually ask
Most operations managers do not struggle to find things to automate. They struggle to choose. Every team has a long list of tedious tasks, a limited amount of time, and pressure to show a result quickly.
The usual mistake is to start with the task that annoys people most. Annoyance is a poor guide. A task that irritates one person for ten minutes a week is a bad first project. A dull task that quietly consumes six hours a week is a good one.
This guide gives you a repeatable way to pick. It uses a simple scoring method, a worked example with illustrative numbers, and a short checklist to run before you commit. You do not need special software to follow it. A spreadsheet is enough.
Why order matters more than the tool
The first automation you build sets the tone for every one after it. If it saves visible time and runs reliably, your team will ask for more. If it breaks or needs constant fixing, people will quietly go back to doing the work by hand.
There is a well-known warning on this point, often attributed to Bill Gates:
"The first rule of any technology used in a business is that automation applied to an efficient operation will magnify the efficiency. The second is that automation applied to an inefficient operation will magnify the inefficiency."
[Owner note: please verify the exact wording and original source of this quotation before publishing, or replace it with a quotation you can cite.]
The practical lesson is to understand and tidy a process before you automate it. A messy process automated is still a messy process, only faster.
Step 1: List every repetitive task
Gather the people who do the work and ask them to name tasks they repeat weekly or daily. Do not filter yet. Write everything down.
For each task, capture three facts:
- What triggers it. For example, a new order, a month-end deadline or an incoming email.
- What the steps are. Aim for five to ten plain-language steps.
- Who does it today. Include how many people touch it.
If a task has no clear trigger or the steps change every time, mark it as unstable. Unstable tasks are poor early candidates.
Step 2: Measure the weekly hours
Estimate how many hours per week the whole team spends on each task. Ask the people who do it, then sanity-check against a typical week. People tend to underestimate small recurring tasks because they are spread across the day.
A rough number is fine. What matters is that you use the same method for every task so the comparison is fair. If two people each spend three hours on a task, the weekly total is six hours.
Step 3: Score how easy each task is to automate
Rate each task from one to five on ease of automation.
- 5: Fixed steps, structured inputs, clear rules, no judgment needed.
- 4: Mostly fixed, with a few predictable exceptions.
- 3: Some judgment is needed, but it can be reduced to rules.
- 2: Frequent exceptions or messy inputs, such as free-form documents.
- 1: Depends on expert judgment, negotiation or relationships.
Be honest here. Optimism at this step is the most common reason automation projects stall.
Step 4: Calculate the priority score
Multiply weekly hours by the ease score. The result is your priority score. A high score means the task is both time-consuming and realistic to automate.
Then apply a risk check before you accept the winner:
- What happens if the automation makes a mistake?
- Can a person review the output before it reaches a customer or a financial record?
- Can you switch it off and return to manual work the same day?
If the answer to the third question is no, pick a different first task.
A worked example
Imagine a hypothetical 12-person distribution company. The operations manager lists four repetitive tasks. All numbers below are illustrative, not drawn from a real business.
| Task | Weekly hours | Ease (1–5) | Priority score |
|---|---|---|---|
| Invoice data entry | 6 | 4 | 24 |
| Weekly sales report | 4 | 4 | 16 |
| Customer onboarding emails | 3 | 5 | 15 |
| Contract review | 5 | 1 | 5 |
Invoice data entry wins with a score of 24. Contract review consumes almost as many hours, but it needs expert judgment, so it scores low and waits.
Now the risk check. Invoice entry touches financial records, so a mistake matters. But the manager can have one person review each batch before it posts, and the team can return to manual entry in a day. That passes the check.
Next, a rough value estimate. Six hours a week across 50 working weeks is 300 hours a year. If the loaded cost of an hour of staff time is $30 (an assumed figure, so substitute your own), that is about $9,000 of annual effort. That figure is the ceiling on what the project can save. Setup and maintenance time must come out of it.
Step 5: Run a 30-day pilot
Do not roll out to everything at once. Run the automation alongside the manual process for 30 days and track three numbers:
- Hours saved compared with your baseline.
- Errors caught during human review.
- Time spent fixing or supervising the automation.
If hours saved clearly exceed supervision time and errors are rare, keep going and move to the next task on your list. If not, adjust the process or drop it. Dropping a weak pilot is a good outcome, because you learned cheaply.
Common mistakes to avoid
- Automating a process nobody has written down. If two people do the task differently, agree on one method first.
- Choosing by annoyance rather than hours. Use the numbers.
- Skipping the review step. Early on, a human should check the output.
- Counting only the savings. Include setup and upkeep when you estimate value.
- Starting with the hardest task. Early wins build trust for harder projects.
Next steps
You can run this method in an afternoon. Build the task list with your team, score each item, and pick the top candidate that passes the risk check. Then pilot it for 30 days and review the numbers honestly.
If you want to know whether working with Patrick Martinchek is a fit for your team, read Who this is for and what it costs.
[Owner note: confirm that /who-this-is-for-and-what-it-costs is the correct path for that page. The business profile provided no details on services or pricing, so none are described here.]