Everyone is talking about AI—on LinkedIn, in podcasts like The Rest is Politics, even in the pub at the weekend. It’s clearly entered the zeitgeist… but you might be thinking: “Should I be using it? And if so, where do I start?”
Sound familiar? You’re not alone.
With Microsoft pushing Copilot, everyone seemingly subscribing to ChatGPT Plus, Google talking Gemini, and Grok being a thing, it’s easy to feel overwhelmed. The first step is to pause and think about your business—what it does and how it operates.
Are you:
- The local chippie selling battered sausages to weather-battered tourists in the Lake District?
- Selling dry robes online via an e-commerce store?
- Running a 400-staff legal firm specialising in commercial property law?
Clearly, each business operates differently, and their AI use cases will differ—but the approach to implementing AI is the same.
Ask the Right Questions
Could what you do—and how you do it—be made simpler or more efficient with AI?
Are there repetitive tasks—like invoicing, drafting, comparing documents, analysing spreadsheets, or even frying chips—that could be automated?
Could AI help you:
- Serve more customers with fewer resources?
- Expand faster while increasing impact or profit?
- Make your staff twice as effective at selling, marketing, advising, or resolving issues?
If the answer is “yes” (or, more likely, probably), your next step is mapping your current processes. Identify gaps, inefficiencies, and duplication of effort.
Map, Rank, and Value Your Processes
Start by mapping processes at a high level. Then:
- Rank the issues by importance.
- Estimate the potential value—what could the business save or gain by improving or automating a process?
- Decide which processes are worth tackling. Small savings (e.g., £5,000/year) may not justify the effort—but beware of knock-on effects elsewhere.
This exercise also highlights:
- Duplication of effort (e.g., multiple teams entering the same data into a system)
- Process loops (where work floats around until someone eventually notices)
Consider a Pilot or Proof of Concept
Once you’ve identified promising processes, try a small-scale AI pilot.
For example: an AI agent could read incoming customer emails, classify them (order, complaint, enquiry, spam), and even start processing them:
- If a client orders 10 red dry robes in medium, AI could verify the client, check SKUs, apply discounts from your ERP system, and draft a sales order for review.
- For a new enquiry, AI could send initial questions to engage the client, help recommend the right product, or provide a “buy now” link.
The result? Significant time savings and faster, more consistent customer service.
Don’t Get Distracted by the Hype
AI can seem technical and intimidating—especially with new versions, platforms, and buzzwords emerging constantly. But here’s the truth: the tool doesn’t matter until you understand where AI can actually help your business.
Without that understanding, any implementation risks being a misfit: users bounce off it, processes fail, and your shiny AI project ends up in the “great bin of failure in the sky.”
Assess the value, map your processes, run small pilots, and plan strategically. Then, when the next wave of weather-battered tourists comes in—or your business faces any recurring challenge—you’ll be ready.
