Your First Business Automation: Where to Start (Not by Buying a Tool)

A business owner once wrote to me: "I want to start with automation. Which tool should I buy?"

It's a natural question. And it's exactly the wrong one.

Because almost everyone asking where to start with their first business automation starts from the end: they buy a tool, sign up for software, watch a video about a system someone else praised. Then they go looking for a task to fit the tool. That's like buying a drill before you've decided where the picture hangs. In this post I want to explain where you actually start, and why the honest answer is sometimes that you don't need automation yet, you just need to fix one process. Starting there saves you months of frustration and money that burns.

Why a tool is the worst place to start

A tool is an answer. And before you have a question, every answer is wasted.

When you buy a tool first, three things happen. First, you bend the business to the tool instead of the other way around: now there's a monthly subscription you feel you have to "justify", so you hunt for uses for it. Second, you learn a whole system for one small problem, and most of its features will never touch you. Third, and worst, you solve the task that's easiest to automate, not the one that hurts most. Those are usually not the same task.

The result is familiar to me: two months later the tool sits open in a tab, nobody logs into it, and the manual work is exactly where it was.

So where do you actually start: listen first, then map

When a business owner drowning in manual work comes to me, I don't ask which tool they want. I ask what their day looks like. Where they copy things by hand from one place to another. What they forget. What keeps them at the desk in the evening.

Then I do one simple thing: I map. You take every repetitive task of the week and put it on a page. Not in your head, on a page. Suddenly you see what you can't see while you're living inside it: the same client who signs up, then gets typed again into a spreadsheet, then gets an email you send yourself, then a reminder you write yourself to call. Four manual actions, the same information, four places.

That map is the part that's actually valuable. The tool, afterward, is almost a technical detail. And sometimes the map shows something else entirely: there aren't enough repetitions to justify automating anything, and what's missing is order in one process. That's a legitimate answer too, and I say it out loud.

The right first task: the most repeated, touching the most

Out of that map, you pick one task to automate first. Not by how annoying it is, but by two tests:

• How often it repeats. Something that happens ten times a day is worth more than something that happens once a month, even if the once-a-month one is more irritating. Automation gives back time in proportion to how many times you repeat it.
• How many other things it touches. A task that sits at a junction, where information flows out to many places, is worth more than an isolated one. When you sort it out, several things downstream get sorted along the way.

Onboarding a new client, for example, almost always passes both tests. It happens over and over, and nearly the whole rest of the process branches out of it. Start there, not from the task that was easiest for you to picture as automated.

The smallest piece that's still connected

Here people panic and imagine a giant project. It's the opposite of what I build.

I build the smallest piece that's still connected to the rest. Not the whole dream, just the one task you mapped, sorted end to end, so the information goes in once and flows onward on its own. Small enough to work in weeks, not months. Connected enough that you feel the relief on the very first day. That's what the work I do looks like in practice.

And this is the important point: the first piece has to be a foundation stone, not an isolated island. That way, when you add the next task, it connects. Over time it becomes one coherent system instead of a pile of tools that don't talk to each other. I wrote about that logic at length in my post on the five signs your business needs one system.

What to do this week

Even with no tool and no budget:

Take a page. For one week, every time you do something manual and repetitive, write it on a line. By the end of the week you'll have what most businesses don't: a map of your repetitive work. Mark the task that repeats the most and touches the most things. That's your first automation, and now, and only now, the question about a tool becomes a good question.

If you want to see what this looks like when it's built right, here are real examples of my work. And if you want help mapping that first task, the short questionnaire is a good place to start, with no commitment and no sales call.