Automation is most valuable when it removes repetitive, predictable work from people who should be spending their time on judgement, relationships or craft. It is least valuable when it is applied to a process nobody fully understands, or one that changes every week. The difference between the two is usually visible before a single workflow is built.
Start with the work, not the tool
It is tempting to begin with a platform — n8n, Make, Zapier — and look for things to connect. A better starting point is a list of tasks your team repeats every week. Ask each person to note what they copy from one system to another, what reminders they send manually, and what reports they rebuild.
You will usually find the same handful of tasks appearing across several people. Those are your candidates.
Five questions to score each candidate
- How often does it happen? Daily and weekly tasks pay back faster than quarterly ones.
- Are the steps predictable? If the same inputs lead to the same actions, it can be automated. If it depends on judgement every time, it may only be partly automatable.
- Where does the data live? Tasks involving tools with good APIs are far easier to automate than those relying on email attachments or paper.
- What happens when it goes wrong? Missing a reminder is inconvenient; sending an incorrect invoice is serious. Higher-risk processes need review steps.
- Who owns it? Every automation needs a person responsible for noticing when it fails and updating it when the process changes.
Good first candidates
- Capturing leads from website forms and ads into your CRM
- Appointment confirmations and reminders
- Creating invoices from completed jobs and sending payment reminders
- Syncing customer details between your CRM, accounting and support tools
- Compiling recurring reports from several data sources
What to avoid at the start
Avoid automating processes that are still being designed, processes with many one-off exceptions, and anything where a silent failure would cause serious harm. These can often be automated later, once the process has settled and you have experience running simpler workflows.
Measure before and after
Before building anything, estimate how long the task takes today and how often errors occur. After launch, check the same numbers. This keeps the conversation grounded in real outcomes and helps decide what to automate next.