The standard AI business case looks something like this. Project spend: £3M. Expected annual saving: £4M through headcount reduction and efficiency gains. Payback period: nine months. ROI on a three-year horizon: strongly positive. Approved.
I've seen hundreds of business cases built on this template across my career in fintech and as an investor in technology companies. The template is familiar because it works for a certain class of investment. It's the same template that justifies a fleet upgrade, a new ERP, or a factory automation project. Cost in, cost saving out, net present value positive, decision made.
The problem is that the template doesn't fit what AI investments actually are. An AI deployment is rarely a substitution of one cost for another. It is a capital formation event that changes the intangible asset base of the business. Evaluating it as if it were a headcount-reduction exercise is like evaluating a brand campaign by its cost per impression — technically measurable, fundamentally missing the point.
An AI deployment is a capital formation event, not an expense substitution. A typical project produces six intangible asset movements. Measuring by cost savings captures one of them. The other five usually matter more for enterprise value.
The AI deployments that produce durable returns have a specific characteristic: they build multiple intangible assets simultaneously. The AI is not the asset. The AI is the instrument that produces the assets.
Consider a customer service AI deployment. The surface-level justification is headcount reduction in the support team. The actual outputs of the deployment are more numerous.
| Output | Intangible asset built |
|---|---|
| The AI platform itself | Technology & Innovation |
| Conversation logs, training data, pattern library | Data & Intelligence |
| Prompts, escalation rules, knowledge base | Organisational Capital |
| Faster response times and 24/7 availability | Customer Capital |
| Remaining team handling complex issues with AI assist | Human Capital (upgraded quality) |
| AI-powered service positioning | Brand & Reputation |
That is six intangible drivers moving from a single project. Measuring the project by cost savings alone captures one of those six movements. It ignores the other five, which in aggregate usually matter more for enterprise value.
The same pattern holds across most meaningful AI deployments. An AI coding assistance rollout isn't primarily a productivity gain — it's a technology asset upgrade that changes what the engineering team can build, and consequently what technology assets the company can produce going forward. An AI sales enablement deployment isn't primarily a cost reduction — it's an upgrade to customer capital (better-served prospects), human capital (more effective salespeople), and data assets (structured records of every sales interaction). An AI research and analysis platform isn't primarily a consulting cost reduction — it's a build of intelligence capability that can compound across every future decision the business makes.
The companies that get sophisticated about this reframe their AI investment cases accordingly. The headcount conversation becomes one line of a multi-dimensional view. The dominant line is usually intangible capital formation — how much of each driver is being built, at what rate, with what expected enterprise value contribution.
When AI investment is evaluated as capital formation, three things change in how companies allocate it.
First, the sequencing of investment changes. Capital formation investments are rarely one-time. They build on themselves. The AI customer service platform generates data that improves the next AI deployment. The coding assistant produces artefacts that train the next generation of the tool. The sales enablement platform builds a dataset that is itself an asset. Companies thinking this way invest in a sequence designed to compound, not in a series of independent projects.
Second, the composition of investment changes. A project that only builds technology capital is weaker than a project that builds technology, data, and organisational capital together. Companies thinking this way design their deployments to capture the full set of outputs — investing in documentation, in knowledge transfer, in data infrastructure, and in the human skills needed to direct the AI. The net investment is higher. The net return, over a multi-year horizon, is substantially higher.
Third, the measurement changes. Cost savings are tracked monthly on the P&L. Intangible capital formation is tracked through a different instrument — a driver-by-driver view of the intangible asset base, updated quarterly, connected to enterprise value through growth accounting. This instrument doesn't exist by default in most companies. It has to be built, or adopted, deliberately.
Two finance teams approve identical £5M AI customer service programmes. Team A evaluates it as headcount substitution; the business case shows £6M annual saving against £5M cost, approved. Team B evaluates it as capital formation: the same headcount line plus expected build of Technology (£8M asset value at year three), Data (£3M), Organisational Capital (£2M), Customer Capital (£4M improvement on retention), and a measured culture-and-brand impact. Team B's investment committee approves with a different mandate — invest in documentation and data infrastructure on top of the platform itself, take a more measured pace on headcount, and produce a quarterly driver dashboard. At year three the cost-saving lines are similar. The enterprise value contribution differs by 4× because Team B captured the full asset formation.
The Opagio 12 framework was designed for this reframing. When a company runs its AI investment cases through the Opagio 12 lens, each proposed project is evaluated for its expected movement across all twelve drivers. The project justification stops being a single ROI number and becomes a portfolio impact projection. The headcount line is still there, but it sits alongside the technology build, the data asset formation, the organisational capital changes, and the customer capital implications. The conversation at the investment committee is different, and the decisions it produces tend to be different.
The Growth Accounting Engine, grounded in the CHS academic framework (Corrado-Hulten-Sichel — adopted by the OECD, the ONS, and the Bank of England), provides the link to enterprise value. It separates the intangible capital contribution from the rest of productivity growth, which in turn connects the driver movements to financial outcomes. A CEO running this framework can answer the question that really matters: how much of our enterprise value growth came from our AI investment, and which drivers contributed most?
The companion pieces in this series make the case from different angles. The measurement layer the AI economy is missing covers why standard reporting doesn't see this. The seven drivers that move when you deploy AI covers the driver-by-driver breakdown. Twenty-five years of automating knowledge work is the inside-the-cockpit view of why this measurement instrument matters.
The AI deployments that will define the next decade aren't the ones with the biggest headcount reduction stories. They are the ones that build intangible capital most effectively. The companies that recognise this early will allocate their AI budgets differently, sequence their investments better, and end up with stronger intangible portfolios than their competitors.
The shift is from AI-as-cost-saving to AI-as-capital-formation. It's a different discipline, it requires a different measurement framework, and the companies practising it are quietly building advantages their competitors won't see until it's too late to match.
If your next investment committee will see an AI business case, the most useful thing you can do beforehand is baseline the intangible portfolio it will affect. Two ways to start:
Headcount reduction is a tactic. Capital formation is the discipline. The companies that get this right will be the case studies in three years.
Every technology wave builds its own measurement discipline. Manufacturing gave us cost accounting. Programmatic marketing gave us attribution. Subscription software gave us ARR. AI is in the gap now — and 90% of firms report zero measurable productivity impact from their AI spending. Here's what the missing measurement layer needs to do.
Read more →
When a CFO reviews an AI investment, the dashboard shows cost savings, implementation budget, and a productivity metric. What it almost never shows is the movement of the intangible asset base — seven drivers typically move simultaneously. Most companies track one.
Read more →
Block cut 40% of its workforce and tied it directly to AI. The market cheered. But a workforce transformation at this scale moves at least five intangible value drivers simultaneously — most companies measure two. Here's what that gap actually costs.
Read more →Get the latest insights on intangible asset growth and productivity delivered to your inbox.
Book a free consultation to see how Opagio Intangibles can help your business.