I talk to a lot of small business owners. Most of them assume that when I bring up AI, we’re about to discuss machine learning models or building proprietary algorithms. Then I tell them what their peers are actually doing with it, and the response is usually: “That’s it? That’s so simple.”
It’s simple because that’s the point. Most small businesses aren’t using AI to do anything fancy. They’re using it to handle the repetitive stuff that shouldn’t need a person. Things that take surprising amounts of time but follow the same pattern every time.
The Actual Work Being Done
I’ve spoken to accountants who use AI to extract data from invoices and receipts. No more manual data entry. No more squinting at blurry photos of receipts. The software reads them, pulls out the amounts and dates, and slots them into the right categories. It saves roughly an hour per day for someone on their team. Over a year, that’s a person’s worth of working time reclaimed.
Email is another big one. One marketing agency I know uses AI to draft initial responses to client inquiries. Their team still writes the final email, but the first draft cuts their response time by half. Most inquiries follow the same pattern anyway, so you might as well let the tool handle the first pass.
Meeting notes is a big one. AI transcribes and summarises. Instead of someone spending forty minutes writing up what was said, a small team gets a two-minute summary with action items pulled out automatically. I know a financial advisor who saves that to her client files. Less admin, better records.
Customer service triage is another one. When inquiries come in, AI sorts them by urgency and topic before they reach a human. Urgent technical issues go to the right person immediately. Password resets get handled by a chatbot. The team spends their time on the problems that actually need a human. One salon owner told me this alone improved her customer satisfaction scores because people weren’t waiting days for responses to simple questions.
And marketing content. Small businesses need a lot of it, social media posts, blog updates, email campaigns. AI can generate first drafts. Not the final copy. The first draft that gets edited, shaped, and made human before it goes out. That’s how I think about it anyway.
What AI Still Can’t Do
AI is terrible at things that require real judgment, nuance, or understanding of your specific business. It can’t make strategic decisions. It can’t replace a conversation with a customer who has a complex problem. It definitely can’t do your taxes properly or give you legal advice, no matter how much it confidently tries to.
It’s also not great at understanding the subtle things. Your brand voice. The specific way your team works. Why certain clients need a phone call instead of an email. That’s all still you.
The Money Part
One thing that surprises people: the tools that do this are cheap. I’m talking under £50 a month for most of them. Some are even free with limits. You’re not investing in expensive infrastructure or hiring data scientists. You’re paying for software that’s already built and ready to use.
Start with the accounting software that has built-in invoice processing. Grab ChatGPT Plus if you need email drafting. Use Otter or Fireflies for meeting notes. Add them up and you’re probably at £30 or £40 a month. The time you get back is worth far more than that.
Where to Actually Start
Pick one thing that’s genuinely painful. Something repetitive that you or someone on your team does at least once a week, and it feels like a waste of their talent. Document processing? Email drafting? Customer inquiry sorting?
Look for an AI tool that handles that one thing. Try it for a month. If it works, it pays for itself. If it doesn’t, you’ve lost nothing. Then move to the next painful thing.
The businesses doing this well aren’t trying to transform their entire operation. They’re just fixing the annoying bits, one at a time.
If you want to talk through what might work for your business, get in touch at /contact.