AI is everywhere at the moment. Most business owners I speak to are either already using it, experimenting with it, or wondering whether they are falling behind by not doing more.
That was certainly reflected in yesterday's Help to Grow: Management Module 2 – AI and Digital Transformation, where Justin Turner led the session as subject matter expert. The module covered business model innovation, AI as an opportunity and disruptor, technology implementation and cyber security.
There was plenty of discussion around AI tools, productivity and where businesses could save time.
But for me, the strongest takeaway was much simpler:
Don't start with AI. Start with the foundations.
Build from the foundations
One of the most useful slides in the session was Justin's "Build from the Foundations" model.
The message was clear: before a business starts layering automation and AI into day-to-day activity, it needs the basics in place first.
That means:
- sound technology fundamentals
- a reliable finance stack
- clear onboarding
- defined people and processes
- and only then, automation and AI
The slide summed it up neatly:
"Strong foundations make everything above work better."
That is probably the best place for most SMEs to start.
Because if your systems are weak, your processes are inconsistent, your roles are unclear or your data is poor, introducing AI will not magically fix the problem.
It may simply make a weak process happen faster.
Understand how the business creates value
Another slide from the session looked at how your business creates value.
The sequence was straightforward:
Inputs → Process → Outputs → Results → Financials
That sounds basic, but it is incredibly useful.
Every business ultimately takes inputs, people, skills, materials, time and money, and converts them through processes into products, services or experiences.
Those outputs create customer results.
And those results eventually show up in the numbers: revenue, profit, cash flow and overall business value.
The important point is that technology should improve that chain, not sit separately from it.
If AI is going to be useful, it should help improve one or more of those stages.
- Perhaps it reduces admin.
- Perhaps it improves forecasting.
- Perhaps it shortens lead times.
- Perhaps it improves customer response times.
- Perhaps it reduces rework or error.
- Perhaps it allows your team to spend more time on higher-value activity.
But the starting point should still be:
Where is value being created today, and where is it being lost?
Define the process before you automate it
This was another key point from the session.
Justin's simple three-stage model was:
- Define the process
- Automate
- Apply AI where appropriate
The logic is difficult to argue with.
You can only automate or use AI properly if you understand what the process is in the first place.
That means mapping out:
- what happens
- who does it
- what information is needed
- where decisions are made
- where delays occur
- what creates errors
- and what the desired outcome actually is
For many SMEs, there is probably more productivity to be gained from fixing basic process issues than from introducing a new AI tool.
And that matters because most owner-led businesses are currently trying to get more from the resources they already have.
Productivity before complexity
The Help to Grow module highlighted a range of ways digital technology can improve a business model including:
- creating new opportunities
- generating new revenue
- improving efficiency and productivity
- enabling change and reducing error
Those are the outcomes that matter.
Not whether you are using the latest software.
Not whether you have the most AI tools.
Not whether your competitors are experimenting with something new.
The real question is:
Does the technology improve the business?
For SMEs, that usually means one of three things:
- sell more
- improve margin
- reduce cost or waste
The session's profitability model made the same point. Technology is useful when it improves the underlying business system, not when it becomes another layer of complexity.
Use the Business Model Canvas properly
We also revisited the Business Model Canvas, which I think remains one of the most useful strategic tools for SMEs.
It forces you to step back and look at the business as a whole:
- customers
- value proposition
- channels
- customer relationships
- key activities
- key resources
- key partners
- cost structure
- revenue streams
The module used the Canvas to explore how digital technology and AI can improve different parts of a business model rather than simply being bolted onto existing activity.
That is a much better way to think about digital transformation.
Rather than asking:
"Where can we use AI?"
Ask:
"Where in our business model do we have friction, inefficiency or missed value?"
Then decide whether technology is the right answer.
AI should solve a clear business problem
This point came through strongly in the session material.
The AI adoption guidance was explicit that AI should be used to solve a clear business problem, improve existing workflows and deliver benefits that justify the cost, risk and governance required.
That is good discipline.
Because there is a danger that businesses adopt AI simply because it is available.
But without a clear objective, good data and a defined process, AI can create wasted time, unnecessary cost and new risks. The session also highlighted that AI should not replace judgement, accountability or proper human review.
So before introducing any new AI tool, I would ask:
- What specific problem are we solving?
- What process does it sit within?
- What does good look like?
- What data does it need?
- Who remains accountable?
- What measurable benefit should it create?
If those questions cannot be answered clearly, the business may not be ready.
Cyber security still matters
Digital transformation also creates risk.
The session reminded us that technology adoption cannot be separated from cyber security, data control and governance.
That is particularly important as more businesses use AI tools and cloud platforms.
The module pointed to practical fundamentals such as:
- MFA
- secure backups
- email filtering
- password management
- staff awareness
- data protection
- clear policies on approved tools
Again, the same principle applies: get the foundations right first.
A practical test for SME leaders
If you are considering further AI or digital investment, I would start with five questions:
- Which process is causing us the most frustration or wasted time?
- Is that process actually defined and consistently followed?
- Could standard automation solve the problem before AI is needed?
- What measurable improvement are we looking for?
- Do we have the systems, data, people and governance in place to support it?
If the answer to those questions is clear, then AI may well have a useful role.
If not, fix the foundations first.
Final thoughts
AI will undoubtedly continue to reshape how SMEs work.
Used well, it can improve productivity, decision-making and customer service.
But it is not a shortcut around poor business fundamentals.
That was probably my strongest reflection from yesterday's session.
Technology should strengthen a good business system — not compensate for a weak one.
Or, to borrow Peter Drucker's well-known line:
"There is nothing so useless as doing efficiently that which should not be done at all."
Perhaps the better question for business leaders is not:
"Where can we use AI?"
But:
"Where does our business create value, and what is getting in the way?"
That is where the real opportunity normally starts.
--------------------------------------------------------------------------------------------------------------------
Copy of the Business Model Canvas below courtesy of the Help to Grow Management Programme.
Should you be interested in learning more regarding the HtGM programme do get in touch (note there are eligibility requirements and only for SME UK registered businesses).