AI consultants have appeared everywhere over the past couple of years. LinkedIn is full of them. Your inbox probably has a few cold emails from them. Every other conference has a panel of them explaining how artificial intelligence will transform your industry.
But if you’re a normal business owner — someone running a manufacturing firm, a trades company, a retail operation — you’re probably wondering: what do these people actually do? And would hiring one make any real difference, or is it just expensive hand-waving?
Fair questions. Let me answer them honestly.
What an AI consultant actually does
At its most basic, an AI consultant looks at how your business works and figures out where artificial intelligence can genuinely help. Not in a vague, “imagine the possibilities” way. In a practical, “this specific task is costing you twenty hours a week and a machine could do it in seconds” way.
The work usually falls into three areas.
Identifying opportunities. They look at your day-to-day operations — the repetitive tasks, the bottlenecks, the bits where someone’s copying data between spreadsheets or manually checking invoices against purchase orders. The goal is to find places where AI would actually save time or money, not just sound impressive.
Building solutions. Once they’ve found the opportunities, they build or integrate the actual tools. That might mean an AI system that reads and sorts incoming emails, something that extracts data from scanned documents, or a chatbot that handles common customer questions so your team doesn’t answer the same thing fifty times a day.
Measuring results. A good consultant doesn’t build something and walk away. They measure whether it’s working — is it saving the time they said it would? Is the accuracy good enough? If something isn’t delivering, they adjust it.
What a good one looks like
Here’s where it gets tricky, because the barrier to entry is basically zero. Anyone can call themselves an AI consultant.
A good one will ask questions before proposing solutions. They should want to understand your business, your processes, and your pain points before suggesting anything. If someone pitches you an AI solution in the first meeting without understanding your operations, they’re selling, not consulting.
They’ll speak in plain English. If they can’t explain what they’re proposing without words like “synergy” or “leveraging machine learning capabilities,” be cautious. Either they don’t understand it well enough to explain simply, or they’re deliberately making it sound more complicated than it is.
They’ll be honest about limitations. AI is very good at certain things — pattern recognition, processing data, handling repetitive tasks — and genuinely rubbish at others. A good consultant will tell you when AI isn’t the answer. If every problem looks like an AI problem to them, they’re a hammer looking for nails.
And they’ll show you real examples. “We built an invoice matching system for a manufacturing client that cut accounts processing time by 70%” is credible. “We help businesses unlock the power of AI” is marketing fluff.
What a bad one looks like
You’ll know them by the buzzwords-to-substance ratio. Heavy on jargon, light on specifics.
They promise transformational results without understanding your business. They push expensive solutions when a simple automation would do. They can’t explain how their proposed solution actually works. They have no portfolio of completed projects, just LinkedIn posts. And they use fear — “your competitors are all doing this” — rather than demonstrating genuine value.
The AI consulting space is a bit like the early days of social media marketing. Lots of big claims, big fees, and very little delivery. The good ones are out there, but you need to filter carefully.
The process: what to expect
A decent engagement typically runs like this.
Audit. They spend time understanding your business — your operations, your systems, your pain points. A couple of hours for a small business, maybe a few days for a larger one.
Recommendations. Based on the audit, they come back with prioritised suggestions. What has the biggest impact for the least effort? What’s a longer-term project? Good consultants are also clear about what they’re not recommending and why.
Build. If you go ahead, they build or integrate the solution. Simpler projects take a week or two; complex ones might take a couple of months. Clear milestones and regular updates throughout.
Measure and refine. After the solution is live, they track performance and make adjustments. AI systems often need tuning — the first version rarely works perfectly. What matters is that it improves over time and the impact is measurable.
When you need one
You’ve got repetitive, time-consuming tasks that eat up staff hours — data entry, document processing, answering the same questions over and over, matching invoices to orders. These are prime candidates.
You’re sitting on data but not getting insights from it. Years of sales data, production data, or customer data in spreadsheets — AI can surface patterns you can’t see manually.
You want to scale operations without hiring proportionally more staff. Automation lets you handle more volume without the overhead.
When you don’t
Sometimes you don’t need one. And here’s the bit most AI consultants won’t tell you.
If your processes are chaotic and undocumented, AI won’t fix that. You need to sort the processes first. Automating a mess just gives you an automated mess.
If your data is still on paper — handwritten job sheets, printed invoices filed in boxes — you’ve got a digitisation problem, not an AI problem. Get the data digital first.
If you’re thinking about AI because everyone’s talking about it and you feel like you should be doing something, that’s not a good enough reason. AI for the sake of AI is expensive and pointless.
And if your budget is very tight, there might be simpler tools — Zapier, Make, or some clever spreadsheet formulas — that get you 80% of the benefit at a fraction of the cost.
Finding the right one
Find someone who’s done real work for businesses like yours. Ask for specific examples and results. Make sure they explain things without jargon. Get a clear scope and price before committing.
We offer AI software and integration services for small and medium businesses, focused on practical solutions that save time and money. No buzzwords, no overselling.
If you want to talk through whether AI could help your specific situation, book a consultation and we’ll give you an honest assessment — including telling you if now’s not the right time.
Frequently asked questions
What does an AI consultant do for small businesses?
They look at how your business operates, identify tasks where AI or automation could save time and money, and build or integrate the right tools. For small businesses, this typically means automating repetitive admin, improving data processing, or handling common customer enquiries automatically.
How much does AI consulting cost?
A simple automation project might cost a few hundred to a couple of thousand pounds. More complex work involving custom AI systems can run into tens of thousands. A good consultant will scope the work and give you a clear price before you commit.
How do I know if my business is ready for AI?
You’re likely ready if you have repetitive tasks that follow clear patterns, your data is digital, and you’ve got documented processes. If your operations are chaotic, your data is mostly on paper, or you’re not sure what problem you’re solving, there’s groundwork to do first.
Is AI worth it for a small business?
It can be, but only if applied to a genuine problem. The businesses that get the most value have specific, repetitive tasks eating up staff time. If someone spends ten hours a week on data entry and an AI tool does it in minutes, that’s a clear return. If you’re just adopting AI because it’s trendy, you’ll probably waste money.