Where AI Actually Adds Value for Small Businesses (And Where It Doesn't)

There's no shortage of AI hype right now. Every second LinkedIn post promises that AI will "transform your business." But for most small and medium businesses, the question isn't whether AI is powerful — it's whether it's powerful for them.

After years of building software for businesses across energy, transport, and IoT, I've developed a practical framework for identifying where AI creates real value. Here's what I've learned.

Where AI Shines for SMBs

1. Repetitive Knowledge Work

If your team spends hours doing the same type of analysis, writing, or data processing, AI can likely do it in minutes. Think:

Summarising customer feedback or support tickets
Drafting initial responses to common enquiries
Extracting structured data from unstructured documents
Generating reports from raw data

The key question: Is someone doing this manually more than 5 hours per week? If yes, there's likely an AI opportunity.

2. Customer-Facing Communication

AI excels at handling the first layer of customer interaction:

Answering FAQs with context-aware responses (not the frustrating chatbots of 2020)
Routing support tickets to the right team
Personalising email campaigns based on customer behaviour
Generating product descriptions at scale

3. Internal Search and Knowledge

Most businesses have critical information scattered across documents, emails, and systems. A well-built RAG (Retrieval-Augmented Generation) system can give your team instant access to institutional knowledge — like having a search engine that actually understands your business.

Where AI Doesn't (Yet) Add Value

1. Decisions That Require Accountability

AI should inform decisions, not make them — especially where compliance, safety, or significant financial risk is involved. A human needs to own the outcome.

2. Small Data Problems

If you don't have meaningful data, AI can't help much. "We want AI to predict customer churn" falls flat when you have 50 customers and no tracking.

3. Problems That Don't Exist Yet

The most common waste I see: businesses building AI solutions for problems they might have, instead of problems they do have. Start with the pain point, not the technology.

A Simple Framework

Before investing in AI, ask three questions:

Is there a clear, repeatable problem? AI needs patterns to be useful.
Do we have (or can we get) the data? No data, no AI.
What's the cost of the current approach? If a manual process costs $500/week in labour, an AI solution needs to save more than it costs to build and maintain.

The Bottom Line

AI isn't magic. It's a tool — a powerful one, but a tool nonetheless. The businesses getting the most from AI aren't the ones with the biggest budgets. They're the ones asking the right questions about where it fits.

If you're not sure where to start, that's exactly what a strategy session is for. I help businesses cut through the noise and find the AI opportunities that actually matter.