What Can AI Do for Business? A Practical, Honest Guide

Last year, a business owner asked me a question I hear every week.

“Mícheál, should I be doing something with AI? I don’t know where to start, and I don’t want to waste money.”

Sound familiar?

I get it. The noise around AI right now is deafening. Every tech company wants to sell you something.

Every headline promises overnight transformation. And meanwhile, you’re running a business with real people, real pressures, and no time for experiments.

Here’s the truth. What AI can do for business is genuinely impressive. But most of the advice out there is written for enterprises with thousands of employees and millions in budget. Not for businesses like yours.

I’m the founder of AI With Kelso, and I work exclusively with businesses that have between 20 and 200 employees in Ireland and the UK. I’m currently enrolled in MIT’s Agentic AI for Organisational Transformation programme. But what matters more than any credential is this: I only take on clients where I can see a clear path to covering their AI investment within six months.

So let me give you a straight answer. No hype. No jargon. Just what I’ve seen work.

What Can AI Actually Do for a Business Like Yours?

Let’s cut through the noise.

For businesses with 20 to 200 employees, AI isn’t about building robots or replacing your team. It’s about removing the manual work that eats your time, your money, and your patience.

Here’s what I see AI doing well for businesses like yours:

Automating repetitive tasks. Data entry between systems. Invoice processing. Report generation. Customer enquiry routing. These are the tasks your team does every week that follow the same pattern. AI handles them faster and more consistently than any person can.

Improving decisions with better data. Most businesses are sitting on data they’re not using. AI can surface patterns in your sales, your operations, and your customer behaviour that would take a human analyst weeks to uncover.

Streamlining internal communication. Meeting summaries, project updates, document drafting. AI won’t replace your team’s judgement, but it takes the admin off their desks so they can focus on work that actually matters.

Enhancing customer experience. Faster response times, personalised follow-ups, consistent service quality. Not chatbots that frustrate people. Intelligent tools that genuinely help.

I worked with a business recently that came to me convinced they needed a custom AI application. Something complex and expensive. After we sat down and mapped their actual processes, the real problem turned out to be a manual step in their invoicing workflow. It was costing them eight hours a week.

We automated it in two weeks. No custom build needed.

AI for business automation doesn’t have to be complicated. Sometimes the biggest win is the simplest one.

How Does AI Help Businesses in Practice? (Real Examples)

This is where I’d rather show you than tell you.

One of my clients, a distribution company with about 60 staff, was drowning in weekly reporting. Every Friday, three people spent the best part of a day pulling data from their CRM, their warehouse system, and their accounts software. All to build a management report for Monday morning.

The report was always late. The data was slightly out of date by the time leadership read it. And the team hated doing it.

Here’s what actually happened when we got involved.

During the Review phase of our AI Advisory process, I sat with the team and mapped exactly how the data moved. Where it came from. What was manual. What broke regularly.

We discovered that 80% of the work was just copying data between systems and reformatting it into a single spreadsheet.

We piloted an automated workflow that pulled data from all three systems overnight, compiled it into a clean dashboard, and delivered it to the leadership team by 7am every Monday. No manual work. No stale data. No Friday scramble.

The outcome? Those three team members got their Fridays back. Leadership had real-time visibility instead of a week-old snapshot. And the entire automation paid for itself within the first month.

But here’s the lesson most people miss.

The biggest win wasn’t the technology. It was removing friction that the team had simply accepted as “just how we do things.” Nobody had questioned the process in years. They assumed it had to be that way.

That’s how artificial intelligence helps businesses in practice. It’s not magic. It’s finding the work that shouldn’t be done manually and taking it off people’s desks.

I’ve seen similar results with customer enquiry routing, onboarding document preparation, and internal knowledge management. The pattern is always the same: find the bottleneck, automate the repetitive part, give people their time back.

Is AI a Good Investment for Small Businesses?

Let’s face it. AI is not cheap. But doing nothing isn’t free either.

According to McKinsey’s State of AI report, 88% of organisations now use AI in at least one business function. And Accenture’s research shows that 74% of organisations report their AI investments have met or exceeded expectations.

But here’s what those numbers don’t tell you. Most of that data comes from large enterprises with dedicated AI teams and deep pockets.

For businesses with 20 to 200 employees, the question isn’t “is AI profitable?” It’s “where do I start so it pays for itself quickly?”

PwC’s research on private companies and AI makes an encouraging point here. They found that AI can be more valuable to smaller companies than larger ones, because it reduces the advantages of scale. In plain English, AI can help your 50-person business operate with the efficiency of a company three times your size.

That’s exactly how I approach every engagement. I don’t take on a client unless I can see a clear path to covering the AI investment within six months.

That’s the AI With Kelso Guarantee. If you implement our AI Growth Strategy as agreed and don’t recover your investment through cost savings and increased revenue within six months, we pay you the difference.

What does it actually cost to get started? It depends on the problem you’re solving. A focused automation project might run a few thousand euros.

But the right question isn’t “what does it cost?” It’s “what is this problem costing you right now?”

I worked with a business that was spending the equivalent of a full-time salary on manual data processing alone. We automated that work for a fraction of the cost. The investment recovered itself within three months, and the savings compound every single month after that.

Is AI a good investment? Yes, if you start with the right problem and measure the return honestly.

What AI Can’t Do (And Where Businesses Get It Wrong)

I’ve seen businesses waste money on AI. And it’s almost always for the same reason.

They bought tools before they understood the problem.

One business I came across had already invested in three different AI platforms before we ever spoke. None of them were being used.

The team didn’t understand how the tools fitted into their daily work, and nobody had taken the time to explain why they mattered. Thousands of euros sitting idle because the adoption step was skipped entirely.

Here’s the thing. That’s the number one mistake I see. Jumping to solutions before defining the problem.

Here’s what else AI can’t do:

AI can’t fix a broken process. If your workflow is messy, automating it just makes the mess faster. You need to understand how work actually flows before you automate anything. That’s why our framework starts with a Review phase, not a technology purchase.

AI can’t replace human judgement. It can inform decisions, surface data, and handle routine tasks. But strategic thinking, relationship building, and leadership still require people. AI is a tool, not a replacement for your team.

AI can’t adopt itself. This is the one that catches most businesses out. You can build the best automation in the world, but if your team doesn’t understand it, trust it, or know how to use it, it gathers dust within a month.

I’ve seen this play out more than once. The technology works perfectly. But adoption fails because nobody invested in training or change management.

That’s why practical AI training, designed around your team’s actual roles and workflows, isn’t optional. It’s the difference between AI that sticks and AI that gets ignored.

How Can AI Help Small Businesses Get Started Today?

If you’ve read this far, you’re probably thinking about your own business.

Good. That’s the whole point.

Here’s my advice. Don’t try to do everything at once. Start with one process. Pick the one that frustrates your team the most, costs you the most time, or creates the most errors.

I use a simple three-step approach with every client:

Step 1: Discover. We start with a free 30-minute AI Discovery Call. I learn about your business, your bottlenecks, and where you want to go. You learn whether AI is the right fit. No commitment. No hard sell. Just a straight conversation.

Step 2: Plan. If there’s a fit, we run an AI Readiness Assessment. This maps your processes, identifies the highest-impact opportunities, and builds a clear roadmap with expected ROI. You know exactly what you’re getting into before anything is built.

Step 3: Implement. We build, test, and deploy. Using our Review, Align, Pilot, Integrate, Deploy framework, we start small, prove the value, then scale what works.

Not sure if you’re ready for a conversation yet? Take the AI Readiness Quiz. It takes two minutes and gives you a clear picture of where AI could make the biggest difference in your business.

The most important thing is to start somewhere. Every week you spend doing manually what AI can handle is a week of time and money you won’t get back. Your competitors are already asking these questions. The cost of doing nothing grows every month.

The Real Question Isn’t “What Can AI Do?” It’s “What Should AI Do for YOUR Business?”

Every business is different.

Your bottlenecks aren’t the same as your competitor’s. Your team isn’t the same. Your processes aren’t the same.

That’s why a blog post can only take you so far. The real answer to what can AI do for your business comes from looking at how your business actually works, where the friction is, and what would happen if you removed it.

The goal isn’t to adopt AI for the sake of it. It’s to scale your profits without scaling your workforce. To take control of your business instead of letting your business run you.

If anything in this post made you think about your own business, that’s exactly what the AI Discovery Call is for. Thirty minutes. No cost. No obligation. Just a straight conversation about where AI fits and whether it’s worth pursuing.

Book your free AI Discovery Call here.