How To Automate Repetitive Tasks In A Small Online Business

TL;DR (Quick Summary)

• Start with the task that's repetitive and doesn't require your expertise, not the most impressive one. The annoying tasks are where the hours are hiding.
• The best automation tools for a small online business are a general AI assistant, your existing email platform's built-in automations, and a connector like Zapier or Make for the gaps. You need fewer tools than you think.
• Automate the repeatable parts of a task, then keep yourself in the loop for the judgment. 
• The tasks worth automating first are usually the invisible ones: finding where a price is typed, keeping a resource list current, turning one video into six assets.
• Never let work reach a client or a student without a person reviewing it first.

How can small online businesses automate repetitive tasks?

Pick one repetitive task that doesn't require your expertise, automate that one properly, then pick the next one.

That sounds too simple to be a strategy. It's the way that works for a business of one or two people, because the alternative is what most people actually do: sign up for six tools in a weekend, half-configure all of them, and end up with more admin and tech issues than they started with.

Here's the sequence I use:

1. Write down what you actually did last week. Not what you meant to do. Track it for at least five days.
2. Find the tasks that repeat. Same shape, different details. Those are the automatable ones.
3. Review the process first. Those 12 steps - could it be 8 steps or does it have to be 12 steps? Make the process as efficient as possible before trying to automate.
4. Split each one into the repeatable part and the judgment part. Gathering the information is repeatable. Deciding what to do with it usually isn't.
5. Automate the repeatable half. Keep yourself on the judgment half.
6. Only then go looking for a tool. Most people do this step first, which is why they end up with eleven half-used subscriptions.

How much time can automation actually save a solopreneur?

Okay, so I tracked one week properly recently instead of estimating, because I'd been loosely guessing and I wanted real numbers.

The work in that week would have taken 51 hours and 25 minutes the old way. It took 4 hours and 30 minutes of my time. That's 46 hours and 55 minutes back across five days, which is insane to think about.

That number needs a caveat, and it's an important one. Let's be honest: most of those tasks would never have been done at all. Lead generation, proper analytics, a searchable library of 116 videos going back to 2017, a correctly configured ad account. Those had been on my someday list for years, and they stayed there because there was never a day with room in it.

Automation didn't take 46 hours of existing work and make it disappear. It made 46 hours of previously impossible work possible. That's a better outcome.

If you want a number that maps to most small businesses: expect to get back somewhere between two and eight hours a week from the first two or three tasks you automate. That's a working week a month, and it compounds, because the time you get back is the time you need to set up the next one.

What are the best automation tools for course creators?

A smaller set than the internet suggests:

• A general AI assistant (Claude, ChatGPT or Gemini) for work involving writing, summarizing, analyzing or restructuring. This is the workhorse and it replaces a surprising number of single-purpose tools.
• Your email platform's built-in automations. Kit, Mailchimp, ActiveCampaign and the rest all have sequence and tagging automation you're probably already paying for and not using. Start here before adding a new tool.
• Your course platform's built-in automations. Thinkific has enrollment triggers, drip schedules and completion actions built in. Most course creators run none of them.
• A connector like Zapier or Make for the gaps between tools that don't talk to each other. Useful, and worth watching, because connectors bill per task and a chatty automation gets expensive fast.
• A transcription tool if you publish video or run recorded calls or group sessions. 

Do I need Zapier to automate my business?

Not to start, and often not at all.

Most small online businesses have more unused automation inside the tools they already pay for than they'd get from adding a connector. Your email platform can tag, segment and send based on behaviour. Your course platform can enrol, drip and trigger.

Your project management tool can template a whole project. Using all of the AI features of the current tools costs nothing extra.

A connector is useful when you have two tools that genuinely can't talk to each other and the handoff between them is costing you real time every week. That's a legitimate reason to add one. "Everyone says you need Zapier" isn't.

What should you never automate?

Work that reaches a client, customer, student or lead without a person reviewing it. My quoting process is now largely automated, and a quote used to take about an hour, because every job is different and the pricing has to be exactly right. It now takes five minutes to produce and five or ten minutes for me to review. I would never send one unread. The automation changed the drafting, not the responsibility.

Your own judgment about what matters. The tool doesn't know which client is difficult, which project is underpriced, or which student needs help. You do.

Relationship touchpoints that people can tell are automated. A birthday email with the wrong name is worse than no birthday email.

A process you don't understand yet. I've spent about $100 on ads in 25 years of business. When I finally set up an ad account properly, I had AI do the research and the setup, because I'm the kind of person who likes to understand the basics before I hire it out. Automating a process you can't evaluate means you can't tell when it breaks.

How do I know if a task is worth automating?

Three questions, and it has to pass all three.

1. Does it repeat? Same process every time, different details. A one-off isn't worth automating no matter how annoying it was.
2. Can you describe it to another person in under five minutes? If you can't explain the steps, you can't hand them off, to a person or a tool. Tasks that fail this one usually need to be simplified before they're automated.
3. Would you notice if the task didn't get done? If the answer is no, don't automate it. Delete it.

That third question kills more candidate automations than the other two combined, and it should. A lot of what fills a small business week is work that made sense two years ago and hasn't been looked at since.

A note on doing this without hiring

I had 28 team members at one point and I let all of them go, because I didn't want to spend my days managing people. I'm not interested in learning to be a good manager. It's not the work I want. I run the business now with one casual contractor and AI, and

I prefer it that way.

That's a real choice with a real tradeoff, and it's worth naming rather than pretending automation is free. Automation replaces the coordination cost of a team with a setup cost you pay yourself, up front, in your own time. For a solopreneur who'd rather do the work than manage the people, that trade is worth making. For a business owner who wants a team, it isn't the same decision.

Where to go next

If you want to find out which parts of your business are already running without you and which parts are still costing you hours, the free AI Opportunity Scorecard takes about two minutes and gives you your score before it asks you for your email.