AI Maturity Scale: Curiosity and Where to Start. A Practical Starting Point for Small and Medium-sized Businesses Exploring AI

The Silent Majority

Despite the constant discussion around AI in business, most SMEs are not adopting AI in a structured way. Many organisations have:

• Experimented with ChatGPT or other Generative AI’s

• Encountered AI features inside existing software

• Heard about automation at industry events

But they typically do not have:

• A defined AI strategy

• Governance or usage policies

• A roadmap for adoption

And this is completely normal. The real problem is not inactivity. It is paralysis created by the way AI is discussed. SMEs are often caught between:

• Enterprise-level transformation messaging

• Concerns around GDPR, compliance and data sovereignty

• Unclear commercial returns

The result is uncertainty about where to begin. This article is for organisations at the AI Curiosity stage of their maturity journey, businesses that are not currently using AI in a deliberate way and want a practical, low-risk starting point.

Why Most Small and Medium-Sized Businesses Haven’t Started

For many leaders, the hesitation around AI adoption is not about interest, It’s about clarity. Common concerns include: “We don’t have the data.” “We are too small.” “AI seems expensive.” “We don’t have the internal expertise.” “What about GDPR and data sovereignty?”

Medium-sized businesses often have additional concerns such as:

• Fragmented data across multiple systems

• Legacy technology environments

• Potential disruption to operational workflows

Another source of confusion is the blurred line between AI, automation, and process improvement. Many organisations assume they need AI when in reality their immediate opportunity may lie in automation or better process design.

The issue is not whether AI is relevant to SMEs. It is that the starting point has rarely been clearly defined for them.

What “Curiosity” Actually Means

When organisations say they are not using, or are curious about AI, what they often mean is that they have not yet adopted it in a structured way.

Typical characteristics of this stage include:

• Manual processes dominating operational workflows

• Heavy reliance on spreadsheets

• ata being repeatedly re-keyed between systems

• Knowledge stored in ‘individuals’ rather than systems

• Repetitive administrative tasks consuming employee time

This stage should not be seen as a failure, In reality it’s simply pre-automation. In many cases the presence of operational friction is actually the clearest signal that a business is ready to begin exploring AI or automation.

Most SMEs are closer to AI readiness than they realise because the opportunity rarely lies in complex technology. It lies in identifying where everyday work is unnecessarily difficult.

The AI Starting Framework

For organisations that are not currently using AI, the starting point should be practical and structured rather than experimental. A simple four step model works well for most organisations.

Step 1: Identify Operational Friction

Instead of asking “How can we use AI?” start by identifying where work repeatedly slows down. Look for areas where teams:

• Repeat the same tasks daily

• Manually process information

• Rely heavily on spreadsheets

• Move data between systems

• Spend significant time producing reports or documentation

In many cases the solution may involve automation or improved process design rather than AI itself. The objective at this stage is simply to understand where the friction exists.

Step 2: Start with Assisted AI

Once opportunities are identified, begin with low-risk, human-assisted AI tools. Examples include:

• Generative AI for drafting internal documentation

• AI-assisted meeting summarisation

• Intelligent features already embedded within existing software

At this stage It’s sensible to avoid:

• Fully automated workflows

• Customer-facing AI deployments

• Complex predictive modelling

• The goal is to build familiarity and confidence before introducing deeper integration.

Step 3: Establish Basic Governance

Even early experimentation should operate within clear boundaries. Organisations should consider:

• Rules around entering client data into AI systems

• Understanding where AI systems store information

• Basic awareness of GDPR implications

• Internal policies governing acceptable use

Governance does not need to be complex, but it does need to exist.

Step 4: Measure Time Saved, Not Innovation

Early AI success rarely appears in revenue figures. Instead, organisations should measure:

• Hours reclaimed Reduction in manual errors

• Faster turnaround times Improved operational consistency

These incremental improvements often compound significantly over time.

Breaking Down the Barriers to Entry

Many SME leaders assume AI adoption requires significant budgets, specialist teams or advanced technical infrastructure. In reality the barriers are often lower than expected. Most organisations already possess:

• Operational data inside existing systems

• Software platforms that contain embedded AI capability

• Cloud environments capable of supporting integrations

The greater challenge is usually lack of structured evaluation. Without structure, experimentation tends to become scattered and inconsistent.

The real risk for many SMEs is not adopting AI. It is allowing teams to experiment with AI tools without clear policies, governance or strategic direction.

This is where structured support can help organisations move from curiosity to readiness in a controlled and safe way.

AI Curiosity in Reality

Organisational size influences where the starting point often lies.

Smaller businesses with 10 to 49 employees typically benefit first from productivity-focused AI tools, particularly generative AI that reduces administrative workload and frees leadership time.

Medium-sized businesses, with 50 to 250 employees, often benefit from process mapping and early automation opportunities. At this stage operational complexity increases and structured workflows begin to matter more.

These organisations should also begin thinking about governance and data policies earlier, as the scale of operations introduces greater regulatory and operational risk.

The Competitive Window

Many SMEs currently sit at the earliest stage of the AI maturity curve. That means the competitive gap has not widened yet. However, it will.

Organisations that begin exploring AI today, even in small and controlled ways, will gradually build operational advantages over the next three to five years.

The businesses that benefit most will not necessarily be those that adopt AI first. They will be the ones that adopt it deliberately.

By starting with operational friction, introducing governance early and scaling capability gradually, organisations can build a foundation that supports long-term efficiency and growth.

The question is therefore not whether businesses should use AI, It’s whether they will approach adoption strategically or allow it to develop accidentally. The next article explores the next stage in the maturity journey.