Learning Velocity: The New Economics of Enterprise Value

I've been thinking about a question that I suspect will become increasingly important for business leaders.

If two companies have access to essentially the same AI technology, what determines which one creates more value?

Capital matters. So do talent, data, customers and market position. But I think there's another factor we're seriously underestimating.

The speed at which a business learns.

I call it learning velocity. And I believe it will become one of the most important determinants of enterprise value over the next decade.

Not because learning has suddenly become important. It always has been.

What's changed is the economics of learning.

The advantage you have today might not be the one you need tomorrow

For most of modern business history, competitive advantage has been about building things that are difficult for others to replicate.

Distribution networks, intellectual property, customer relationships, technology infrastructure and institutional knowledge.

These advantages take time and money to establish. Once you've built them, they can protect your business for years, sometimes decades.

AI is starting to change that equation.

Capabilities that once required years of development and significant investment can increasingly be created or replicated in a fraction of the time.

Of course, not every advantage is disappearing. Trust, relationships, physical infrastructure and regulatory barriers still matter enormously.

But some competitive advantages are becoming less durable.

And that raises an interesting question.

What happens when your ability to create the next competitive advantage becomes more valuable than the one you already have?

I think that's the economic shift we're beginning to see.

Learning is becoming a compounding advantage

When I talk about learning velocity, I don't mean how quickly your employees complete an AI training course.

I mean how quickly your organisation can take new information, turn it into better decisions, change what it does and create measurable value.

There's a big difference.

I've worked with organisations doing an extraordinary amount of AI activity. Training programs, pilots, workshops, productivity initiatives.

They're busy. Sometimes impressively so.

But ask what's actually changed about how the business operates, makes decisions or creates value, and the answers become less convincing.

Activity isn't learning.

Learning happens when new information changes what you believe and what you do.

And here's where things get interesting.

Learning can compound.

SenseReasonActLearn

Sense · Reason · Act · Learn

Imagine two businesses in the same industry, with similar resources and access to the same AI technology.

The first uses AI to improve productivity. It automates processes, reduces costs and makes its operations more efficient.

Good investments. Real returns.

The second does those things too, but goes further.

It uses AI to understand customers differently, challenge assumptions, experiment with new services and discover opportunities it couldn't previously see.

More importantly, it changes how the organisation operates so it can act on what it discovers.

Every experiment generates evidence. That evidence improves the next decision. Better decisions create new opportunities to experiment.

Over time, the organisation gets better at the process itself.

The first business becomes more efficient at doing what it already does.

The second develops an increasing capacity to discover what it should do next.

Initially, the financial difference might be negligible. Over several years, those differences can accumulate.

That's the compounding advantage.

And it's much harder to copy than a piece of software.

Follow the money

This is where learning velocity becomes a serious boardroom conversation.

Enterprise value is fundamentally about expectations of future cash flows, growth and risk.

Now consider what happens when a business becomes consistently better at learning.

It can identify customer opportunities earlier, test propositions before committing significant capital, recognise failing investments sooner and discover new sources of revenue.

Each of these affects the underlying economics of the business.

Learning velocity

1. Opportunities identified earlier
2. Propositions tested before capital is committed
3. Failing investments recognised sooner
4. New sources of revenue
5. Better capital allocation
6. Lower risk

And we're beginning to see evidence of this distinction in how organisations are capturing value from AI.

BCG's 2025 research found that the 5% of companies most advanced in their AI capabilities achieved 1.7 times the revenue growth and 2.7 times the return on invested capital of less mature organisations.

Perhaps more strikingly, they also recorded 3.6 times the total shareholder return over three years.

These companies were doing something different. They were redesigning how their businesses operated, developing new capabilities and reinvesting the value they created into further growth. BCG, 2025.

McKinsey's research tells a similar story. Its highest-performing AI organisations were more likely to redesign workflows and pursue innovation and growth, rather than concentrating primarily on efficiency. McKinsey, 2025.

Neither study proves that learning velocity is responsible for these financial outcomes.

But the pattern is worth paying attention to.

The organisations creating the greatest value aren't simply using more AI. They're changing how they operate and what they're capable of doing.

And there's a broader economic foundation for this argument.

Research by economists Nicholas Bloom and John Van Reenen found that better management practices were associated with stronger productivity, profitability, sales growth, company valuations and survival rates.

That research predates the generative AI revolution by nearly two decades.

The relationship between organisational capability and financial performance isn't new. What's new is the opportunity AI creates to dramatically accelerate how those capabilities develop.

Which brings me to what I think is the most important point.

A business that learns faster may be better positioned to discover and capture opportunities that aren't even visible today.

That's potentially enormously valuable.

Because investors aren't simply buying the earnings a business generates now. They're placing a value on what they believe it can generate in the future.

If one business can repeatedly discover new sources of growth while another depends on extending its existing model, why would we expect their future values to be the same?

I don't think we should.

The uncomfortable leadership question

There's a problem, though.

Most organisations weren't designed to learn particularly quickly.

They were designed to deliver predictable outcomes, manage risk and optimise existing operations.

Their structures, budgets, incentives and approval processes reflect that.

Which is understandable. These things exist for good reasons.

But AI creates an interesting tension.

We can now analyse possibilities, challenge assumptions and generate insights at extraordinary speed.

Yet many organisations still take months to decide what to do with that information.

The technology is getting faster. The organisation isn't.

And that's where I think the greatest risk lies.

We're seeing businesses invest heavily in AI adoption without fundamentally changing how they operate.

They measure how many employees use Copilot, how much time people save and how many processes have been automated.

Useful measures, certainly.

But they tell us very little about whether the business is developing the capability to create its next competitive advantage.

You can have thousands of employees using AI every day and still have an organisation that's incapable of changing its mind.

That's a leadership problem.

It raises questions about who gets to experiment, how quickly evidence can challenge an established assumption, and whether people have permission to act on what they discover.

It also requires leaders to become more comfortable with uncertainty.

Because if you're serious about learning, you have to accept that some of your current assumptions are wrong.

And that your organisation needs to be able to discover which ones.

The next enterprise value equation

I think we're approaching a point where business leaders need to reconsider what makes their organisations valuable.

The assets and advantages they've accumulated still matter.

But so does the speed at which they can renew those advantages.

That means boards should start looking beyond traditional AI adoption measures.

1. 01How long does it take us to move from identifying an opportunity to testing it?
2. 02How quickly do we recognise when an assumption is wrong?
3. 03How often does new evidence actually change a decision?
4. 04Are we improving our ability to learn over time?

These questions won't replace revenue, margin or return on invested capital. But they could help explain how an organisation will sustain and improve those outcomes.

And there's a more confronting implication.

If learning velocity compounds, then so can the competitive disadvantage of organisations that fail to develop it.

A competitor that learns faster can identify opportunities sooner, improve its products more frequently and allocate capital more effectively. If it reinvests those gains in its ability to learn, the distance between the two businesses can grow.

The slower business may remain profitable for years. It might even look successful against its historical benchmarks.

Until the market changes enough that those benchmarks become irrelevant.

Discovers its next value

More efficient today

I suspect we'll look back at this period and discover that many businesses spent enormous amounts of money using AI to optimise operations that were already losing their relevance.

Others will have used the same technology to build organisations capable of continually discovering where their future value comes from.

And that's the distinction I think leaders need to understand.

The businesses that succeed won't necessarily be those spending the most on AI.

They'll be the ones that learn how to turn intelligence into action, action into learning, and learning into an increasing capacity to create value.

Which leaves a question worth taking into your next board meeting.

If your competitors had access to exactly the same technology tomorrow, what would make your business more valuable five years from now?