The fear is real. It's just pointed at the wrong thing.

Almost every business owner I've talked to this year has asked me some version of the same question.

Female business owner looking worried — ***Is this going to make me unnecessary?***

It's a fair question. The headlines have been relentless, and most of them are written to alarm rather than inform. But I've noticed something about the people asking it. They are never sitting still. They're answering emails at 9pm. Writing their own service pages between clients. They are the bookkeeper, the marketer, the scheduler, and the person who actually does the work.

They are not at risk of being replaced. They are at risk of burning out.

That gap — between the fear and the actual situation — is what I want to talk about.

Replacement is the wrong frame for a business of one.

When people imagine AI taking a job, they're picturing a large organization. A role on an org chart, a set of tasks, someone in a seat, and a system that can do the seat's work more cheaply. That's a real conversation in some industries, and I won't pretend otherwise.

But that's not the shape of the businesses I work with. A personal trainer, a hairdresser, a two-person pressure-washing company — these aren't org charts. There is no seat to eliminate. The owner is the business. The relationships are the business. The judgment about which client to take, and how to handle the difficult one, is the business.

Nobody is automating that. What can be automated is the pile of work sitting between the owner and the thing they're actually good at.

The threat to a small business isn't a machine that can do your job. It's a week that never lets you do it.

A better question

So I'd offer a swap. Instead of "will AI replace me," try:

Which three hours of my week would I most like back — and what would I do with them?

That question is smaller, and far more useful. It's answerable. It points at something you can act on this month. And it tends to expose the real problem, which usually isn't technology at all — it's that nobody has ever looked at where the week actually goes.

Here's how I'd run it.

Write down last week. Honestly. Not the version where you were productive. The real one. Include the twenty minutes reformatting a document, the hour rewriting an email you'd already written three times, the Saturday morning catching up on invoicing.

Mark each item with one letter. J for the things that needed your judgment — a client conversation, a pricing call, deciding what to say no to. R for the things that were repetitive — reformatting, summarising, first drafts, admin, data entry.

Add up the R hours. For most owners I've done this with, it lands somewhere between four and ten hours a week. That is the number worth reacting to. Not the headlines.

What I actually hand over, and what I don't

I'll be specific, because vague advice about "using AI" helps nobody.

I use it for first drafts. Not final anything. Starting from a blank page costs me more time than any other single thing in my week, and a rough first pass that I then rewrite completely still saves me an hour. The thinking stays mine. The blank page doesn't.

I use it for summarising. Long reports, meeting notes, the forty-page strategy document a client sends over the night before a call. I need the four points that matter, and I need them before we talk.

I use it for structure. When I have messy notes and can't see the shape, getting them sorted into an outline I can then argue with is genuinely faster than staring at them.

And here's what I don't hand over. I don't let it talk to clients. I don't let it make a recommendation I haven't reasoned through myself. I don't publish anything in my own name that I haven't rewritten in my own voice — because the whole point of my writing is that it sounds like a person who has actually sat across from a business owner and listened. And I'm careful with anything involving confidential client material or community-sensitive context, because that judgment isn't a task, and it isn't mine to delegate.

Happy business owner on the phone — *AI can help me get to the work.*

*It doesn't get to do the work.*

The part nobody says out loud

Getting hours back is only useful if you've decided what they're for.

This is where I watch it go sideways. Someone saves four hours a week and quietly fills them with more admin, or with anxious busywork, or with nothing they can name. Six weeks later they'll tell you the tools didn't change anything. They're right — but not for the reason they think.

So decide first. Before you optimise a single hour, write down what you'd do with three of them. Take on one more client. Finally fix the service page that's been wrong for a year. Follow up with the six people who enquired and never heard back. Leave at five on a Friday.

Any of those are legitimate. Nothing is not.

Where this leaves you

The honest position, as far as I can see it: AI is not coming for the judgment, the relationships, or the accountability that make a small business work. Those were never the parts you could hand off. What it can take is the layer of repetitive work that's been quietly eating your evenings — and for a lot of owners, that's the most valuable trade available right now.

You don't need a strategy. You need an answer to one question.

Which three hours would you take back, and what are they for?

Start there.

If your website is one of those things that's been on the list for a year — the page you know isn't right but haven't had the hours to fix — that's worth a proper look. I offer a free SEO audit where I go through your site the way a ready-to-buy customer would, and send you the few things most worth fixing first. No jargon, no pitch, just a clear starting point.

The Question Isn't Whether AI Replaces You. It's What You'd Do With the Time.