You can automate far more of your business than you realize

Ask a business owner what they'd automate and you'll usually hear the big, obvious things — invoicing, payroll, maybe a chatbot on the website. Those are fine. But the work that quietly eats a week rarely makes that list, because it's buried inside habits nobody questions anymore. Nobody questions them precisely because they're small. A task that takes four minutes doesn't trigger any alarms. A task that takes four minutes, eleven times a day, five days a week, is a part-time job you're paying a senior person to do.

That's the thing about hidden work: it doesn't feel like work you could hand off. It feels like how the job is done.

The test for "is this automatable?"

Here's the rule I use on every call:

If you can describe the steps out loud, it can probably be automated — and it usually should be.

That covers far more than people expect. Copying a lead from an email into a spreadsheet, then into the CRM. Renaming files to match a naming convention before they go in the shared drive. Checking a dashboard every morning and pasting the numbers into a Monday report. Sending the same three follow-ups in a slightly different order depending on whether the client replied. Pulling a list, filtering it, and forwarding the rows that match.

None of these feel like "automation problems." They feel like judgment, or diligence, or just being on top of things. And sure, some of them need a human in the loop. But the mechanics around the decision — the fetching, the copying, the formatting, the chasing — almost never do. That's the part you give back.

Where the hidden work lives

When I audit a process, I'm really hunting for four tells. Once you know them, you start seeing them everywhere in your own week.

• Re-typing — the same piece of data keyed into more than one tool. A lead's name and email living in your inbox, your spreadsheet, and your CRM means a human typed it at least twice.
• Watching — a person checking something on a schedule, just in case it changed. Refreshing a payment dashboard. Glancing at a shared inbox. Looking to see if a file landed yet.
• Chasing — manual follow-ups that depend on someone remembering. The unsigned contract, the unpaid invoice, the document a client still hasn't uploaded.
• Stitching — reports and handoffs assembled by hand from several sources. Three exports, a copy-paste, a bit of spreadsheet math, and a "here's this week's numbers" email.

Each one is a small tax. Individually they're easy to wave off. Added up across a team, they routinely come to a full day per person per week — a day nobody scheduled, nobody planned for, and nobody can point to at the end of it.

A few I've actually pulled out of the floor

Abstract rules are easy to nod along to, so here are concrete ones. None of these were anyone's idea of an "automation project." They were just friction people had stopped noticing.

The agency that re-typed every lead. A 12-person marketing studio captured leads through a website form that emailed them. From there, someone read the email, typed the details into a Google Sheet for tracking, then typed them again into the CRM, then pinged the right account manager on Slack. Three manual steps and a 40-minute average lag before anyone followed up — long enough that warmer leads sometimes went cold. One workflow now does the whole thing the moment the form is submitted: validate the input, create the CRM record, and post to the right Slack channel with a "claim this lead" button. Re-typing gone, lag down to seconds, and the spreadsheet they thought they needed turned out to be a worse copy of the CRM.

The bookkeeper who watched for payments. A small services firm had someone log into the payment dashboard a few times a day to see which invoices had cleared, then mark them paid in the accounting tool and email the client a receipt. Pure watching and re-typing. Replaced with a webhook: the payment provider tells the system the moment money lands, the invoice gets marked paid automatically, and a branded receipt goes out without anyone opening a browser. Nobody refreshes a dashboard anymore — the dashboard tells them.

The onboarding that lived in someone's memory. A consultancy's new-client onboarding was a mental checklist one person carried: send the welcome email, create the shared folder, chase the signed agreement, nudge for the intake questionnaire, set up the kickoff call. It mostly worked — until that person was on holiday, and a new client spent a week in silence. We turned the checklist into a workflow that fires on a signed contract: folder created, welcome email sent, and the questionnaire chased automatically on day two and day five until it's filled in. The chasing — the part most likely to be forgotten on a busy week — is now the part that never gets forgotten.

The Monday report nobody wanted to build. A founder spent the first hour of every week stitching a status report: pull numbers from the CRM, the support tool, and Stripe, drop them into a template, write two lines of commentary, send. The pulling and formatting was the boring 50 minutes; the commentary was the valuable 10. Now a scheduled workflow assembles the figures into the same template every Monday at 7 a.m. and leaves it as a draft. He still writes the two lines that need a human — he just doesn't do the data-entry warm-up act first.

Notice what these have in common. None replaced a person's judgment. They removed the mechanical scaffolding around the judgment — and that scaffolding was where almost all the time was going.

Start small, build it properly

You don't need a platform migration or a six-month project to fix any of this. A single workflow that moves a lead from your form to your CRM and pings the right person is a real, measurable win, and it's the kind of thing you can ship in days, not months. Start with one task. Pick the most boring, most repetitive thing you do every week and make that one disappear first.

The part that matters is building it like real software, not like a clever one-off that only its author understands:

trigger  ->  validate  ->  transform  ->  act  ->  log + alert on failure

That last step is the whole difference between automation you trust and automation that silently breaks at 2 a.m. and quietly stops sending receipts for a week before anyone notices. Error handling, monitoring, and a clear owner aren't extras bolted on at the end — they're what turns a neat trick into infrastructure you can actually depend on. An automation you have to babysit isn't saving you anything; it's just moved the work, not removed it.

The takeaway

You almost certainly have more automatable work than you think, and most of it is invisible to you for the same reason it's invisible to everyone — it's small, it's routine, and it's been "just how we do it" for so long that it stopped registering as a task at all.

The goal isn't to automate everything at once. It's to find the highest-leverage, most boring thing you do every week and make it vanish, then do the next one. Done a few times, that's not a productivity hack — it's a quietly different kind of business, where your best people spend their hours on the work that actually needs a human.

If you want a second pair of eyes on where yours are hiding, that's exactly what an intro call is for.