From AI Investment to Organisational Capability

Welcome to another edition of the Connect blog.

I have been spending a lot of time recently looking at what is happening with AI inside organisations, and I want to share what I am seeing, because I think it matters to everyone reading this, whether you are leading AI transformation, contributing to it as a practitioner, sponsoring it, or responsible for making sure the investment delivers. Some of you are already well into this and seeing results. If that is you, I would love to hear what is working.

TL;DR

Most organisations are investing in AI, but very few are turning that investment into genuine organisational capability. Here is what I cover in this edition:

• What is happening right now

• The numbers behind what I am seeing

• Why this is an organisational challenge, not a technology one

• What this means for leaders and executives

• What this means for transformation and change professionals

• A question to sit with

What is happening right now

Organisations everywhere are rolling out AI tools, and that is a good thing. People are using them to work faster, draft content, summarise documents, and automate routine tasks. Productivity is improving at the individual level, and that is genuinely valuable. For many organisations, this is the first tangible return on AI investment, and it deserves recognition.

But there is a growing gap between what is happening at the individual productivity level and what is needed for AI to become a genuine organisational capability. And that distance is where much of the value is left on the table.

When I say organisational capability, I mean AI embedded in how the business actually operates. Into its processes, its governance, its workforce design, its measurement, its decision-making architecture. Not as an add-on to existing ways of working, but as something that genuinely changes how the organisation creates value for its customers, its shareholders, and its people.

The organisations that have done this well, and there are some, are seeing a compounding effect. AI is not just making individuals faster; it is changing how the business competes, serves customers, and makes decisions. That is a different order of value from individual productivity gains, and it requires a different approach to get there.

Rolling out a tool is a project. Building a capability is a transformation. Most organisations are still treating AI like a project.

The numbers behind what I am seeing

The research supports what many of you are probably experiencing firsthand, and the numbers are striking.

88% of AI proofs of concept never reach production, according to research from IDC and CIO (2025). This is not because the technology did not work. In most cases, the model performed as expected. What was not structured was the organisational side: the capability architecture, the governance, the adoption planning, the workforce transition, the measurement framework that connects to business outcomes rather than deployment metrics.

42% of companies abandoned most of their AI initiatives in 2025, as reported by S&P Global. Not technology failure. Organisational investment that was scattered, governance that was reactive rather than designed, and no strategic architecture connecting individual initiatives to a broader capability agenda.

Only 7% of enterprises say their data is completely ready for AI deployment, according to Cloudera and Harvard Business Review (2025). And yet the pressure to move forward is intense, as the market is moving and no one wants to be seen as falling behind.

These numbers are not a criticism of anyone's efforts. They reflect the reality that AI transformation is a fundamentally different type of organisational challenge, and most approaches used are not structured for it.

The technology is not the bottleneck. The organisational design around it is.

Why this is an organisational challenge, not a technology one

Here is where I think the conversation needs to shift.

In most organisations right now, AI strategy and direction are being led through a technology lens. The vendors are at the table, the data science teams are designing the approach, and the budget often sits with the CTO or a newly created AI function. The conversation is about models, tools, platforms, and implementation timelines.

None of that is wrong. Technology leadership is essential. Vendor expertise is valuable. Data science capability is critical.

What concerns me is what is often missing from the conversation. The organisational dimensions that determine whether AI investment actually becomes embedded capability:

• Governance that is proportionate and enables rather than blocks.

• Adoption planning that goes beyond training to address the deeper concerns people have about what AI means for their roles and their expertise.

• Workforce transition planning that is proactive rather than reactive.

• Measurement that connects to business outcomes, not just how many people are using the tool.

These are not nice-to-haves. They are the dimensions that distinguish organisations where AI is creating genuine value from those where it is stuck at the proof-of-concept stage.

If the organisational side is not designed with the same rigour as the technical side, the investment will not land.

What this means, if you are a leader or executive

If you are accountable for the return on AI investment, the question worth asking is whether the investment is building a lasting organisational capability or creating a collection of individual productivity improvements and isolated departmental initiatives.

Both have value. But they are different outcomes, and they require different investment approaches.

Building organisational capability means thinking about AI the way you think about any other major strategic capability: with governance, measurement, workforce design, and a clear connection to business outcomes. It means making sure the organisational side of AI gets the same level of design and investment as the technical side.

The organisations that are getting this right are those where someone holds the full picture: technology and business, deployment and adoption, efficiency and capability. That is a transformation leadership role, and it is one that needs to be intentional, not accidental.

The question is not whether you are investing in AI. The question is whether you are investing in the organisational capability to make AI work.

What this means if you are a transformation or change professional

The skills you have built over your career, capability design, governance architecture, stakeholder alignment, adoption planning, resistance management, program structure, and workforce transition, are exactly the skills that AI transformation needs and is not getting enough of.

These are not peripheral to AI transformation. They are the core of what determines whether AI investment creates lasting value.

The gap between your existing expertise and AI transformation leadership is not a career restart. It is a specific methodology layer that sits on top of what you already know. How to assess where AI creates genuine business value. How to structure a proof of concept as a capability test rather than a technology demo. How to design governance that enables momentum rather than blocking it. How to measure success beyond deployment metrics.

That methodology layer is learnable, and your existing expertise is the foundation it builds on. The transformation and change professionals who add this layer to their practice in the next six to twelve months will be the ones the market comes to when organisations realise that the technology side alone is not enough.

Your transformation expertise is not becoming less relevant. An AI-Transformation-specific methodology is the missing piece that most AI programs lack.

Where am I going with this

I have been spending the past few months developing a methodology specifically for this: turning AI investment into genuine organisational capability. It is built for experienced transformation and change professionals, and it draws on what I have learned working with global organisations that are seeing clear, measurable results from how they have approached AI.

I will share more about what I am finding in this blog and across my content over the coming weeks. My goal is to give you practical thinking you can apply immediately, whether you engage with anything I offer.

Practical means you can take your expertise into a conversation next week, and it changes the outcome. That is the bar I am holding myself to.

A question to sit with

In the organisation or client environment you work in right now, is AI investment building a genuine organisational capability, or is it a collection of individual tools and isolated initiatives that have not yet been connected to a bigger picture?

One thing worth doing this week:ask someone in your organisation's AI program how success is being measured. If the answer is only about adoption and deployment, the conversation about business outcomes has not happened yet.

If you have a view on this, I would genuinely love to hear it. Send me an email and tell me what you are seeing. I read every reply.