What should I automate first in my business?

Question

VX-N · Accepted Answer

Automate the step that is repeated most often and fails most expensively when someone forgets it — usually lead follow-up, quoting, or invoice chasing. Do not start with the most interesting problem; start with the one that costs money every week it stays manual. Most stalled automation projects stall at exactly this question — not because the tools are hard, but because nobody has written down where the time actually goes. The fix costs a week, not a budget: have each person list what they did more than three times, and beside each item, what happens when it gets skipped. The list that comes back is the roadmap, and it almost never matches what the owner guessed. A useful test: if the task has a clear trigger, a clear rule, and a clear output, it automates cleanly. If it needs judgment on every instance, it does not — and pretending otherwise is how companies end up with automation they have to babysit. The middle ground is the pattern that actually works in production: the system does the mechanical part and drafts the judgment part, and a person approves in seconds what used to take minutes to produce. Sequence by payback, not by ambition. First candidates are the ones where the failure is visible in revenue — a lead that waited two days, an invoice that never went out, a quote that took a week. Second wave is internal drag: data re-entry between systems, status updates someone compiles by hand, reports assembled from exports. Save the clever ideas — the chatbot, the AI agent, the predictive anything — for after the boring wins have paid for them and taught the team to trust the machinery. There is also a real reason to start small that has nothing to do with risk: the first automation teaches your team what a trigger, an exception, and a failure notification feel like. A company that has run one dumb, reliable workflow for three months makes far better decisions about the ambitious one than a company starting cold. The first project's real deliverable is organizational, and choosing an unglamorous, high-frequency task is what makes that lesson cheap.