Every major technology wave has produced its own measurement discipline. When manufacturing industrialised, we built cost accounting. When marketing became programmatic, we built attribution modelling. When software moved to subscription, we built ARR, cohort analysis, and net revenue retention. Each wave created operational decisions that the existing measurement tools could not support, and the measurement tools that filled the gap became the basis for how winners were identified and losers were understood.
The AI economy is in that gap now. Companies are making investment decisions at unprecedented scale — replacing workforces, redesigning operating models, reallocating billions of dollars in annual spend — with a measurement vocabulary built for a pre-AI world. The consequences of the gap are already visible in the research.
These statistics are not evidence that AI doesn't work. They are evidence that companies lack the instruments to tell whether it is working.
The AI ROI problem isn't an AI problem. It is a measurement problem. The instrument that can answer "did this AI investment create value?" does not exist in the standard CFO toolkit — yet.
When I was building automated market data, KYC, and straight-through processing systems in fintech, we had the benefit of working in a domain where productivity was unambiguous. A trade either settled correctly or it didn't. A KYC check either cleared in the required time or it didn't. An FX position was either hedged or it wasn't. The measurement architecture was baked into the operational domain. Whether a given automation was creating value could be answered in milliseconds.
Knowledge work doesn't work that way. When an AI agent drafts a contract, you can measure the time saved but not easily the quality of the drafting versus what a junior associate would have produced. When an AI system handles a customer query, you can measure resolution rate but not the long-term effect on customer lifetime value. When engineering teams deploy coding agents, you can measure lines of code generated but not whether the resulting system is more or less maintainable over a three-year horizon.
This measurement gap is not an accident. It is the direct consequence of a more fundamental problem: accounting standards treat the investments that produce intangible outcomes as period expenses, not as capital formation. Under IAS 38 and FRS 102, most internally generated intangible assets cannot be recognised on the balance sheet. The result is that the majority of what most modern companies are actually building — technology platforms, data assets, customer relationships, organisational capability, brand equity — is invisible to the financial reporting system. The measurement architecture that evolved alongside industrial capital is the wrong instrument for measuring knowledge capital.
AI is the accelerant that makes this mismatch impossible to ignore. A company investing £10M a year in AI agents, automation infrastructure, and the human expertise to deploy them is making one of the largest intangible capital investments in its history. The P&L treats it as operating expense. The balance sheet doesn't see it. The management reporting sees some leading indicators — adoption rates, time savings, cost reductions — but not the asset formation itself. There is no instrument in the standard CFO toolkit that answers the question "how much intangible capital have we built with this investment, and what is it worth?"
A functioning measurement layer for the AI economy has to do four things that the current infrastructure doesn't.
This sounds obvious but is routinely skipped. A £5M investment in an AI customer service platform doesn't just build "AI capability." It builds a technology asset (the platform itself), a data asset (the conversation logs and training data), an organisational capital asset (the processes and playbooks that teach the AI what to do), a customer capital asset (improved response times, personalised interactions), and potentially a human capital asset (the new skills in the remaining team). A measurement framework that only tracks the technology line misses 80% of the asset formation.
The gap between book value and enterprise value in knowledge-intensive businesses is almost entirely intangible. For the measurement layer to matter, it has to quantify how intangible investments translate into enterprise value growth. Growth accounting — the academic framework pioneered by Corrado, Hulten, and Sichel and adopted by the OECD, the ONS, and the Bank of England — does exactly this. It separates the intangible capital contribution from the rest of productivity growth, giving a defensible link between investment and value.
A one-off intangible asset audit is useful for a specific transaction. What companies deploying AI need is a quarterly dashboard that shows the direction of travel — which drivers are strengthening, which are weakening, and how investment is flowing. The parallel is with how CFOs track cash flow: not as an annual exercise but as a continuous view of where the company's financial health is moving.
A senior team making capital allocation decisions across R&D, brand, technology, talent, and data needs to see the relative returns being generated by each category. Currently this is impossible because the investments are buried in operating expenses with no structure to decompose them. Mapping general ledger data to the Opagio 12 categories is designed for exactly this problem.
Imagine a growth-stage B2B software company with £30M in revenue and £8M in annual operating expenses. The CFO can tell you that £2M of the £8M is R&D, £1.5M is sales and marketing, £500K is customer success. That is the P&L view. The measurement layer view is different: £2M of R&D is building a technology asset currently valued at £6.5M on replacement cost, up from £5.2M last year. £1.5M of sales and marketing is building a brand asset worth £3.1M and a customer capital asset worth £4.4M. £500K in customer success is being consumed in operational support rather than building durable relationships — and that is driving the higher-than-sector churn.
When the measurement layer exists, the boardroom conversation changes in three specific ways.
Capital allocation becomes a portfolio discussion, not a line-item debate. The question stops being "can we cut the marketing budget by 12%?" and starts being "are we building the brand and customer capital assets fast enough to defend our enterprise value over the next three years?" The drivers are the unit of debate, not the P&L lines.
AI investment gets scoped by asset formation, not cost savings. The CFO can ask a business unit proposing a £5M AI programme: which of the twelve drivers will this investment build, in what proportion, and what is the expected return on each? The business case becomes a multi-dimensional view, not a single ROI number.
Board reporting becomes integrated. Financial statements sit alongside the intangible dashboard. The CFO presents both. Analysts and investors see the full asset base of the company, not just the portion the accounting standards recognise.
Opagio was built as the measurement layer for this gap. The Opagio 12 framework identifies twelve categories of intangible value, spanning technology, brand, customer capital, human capital, organisational capital, data, and the more recently defined digital-era drivers. The Intelligent Onboarder scans a company's public footprint and financial data to build an initial Value Drivers Register automatically — assembling in minutes what would otherwise be a multi-week manual exercise. The Growth Accounting Engine maps general ledger data to intangible investment categories, producing a Normalised P&L that separates the intangible capital contribution from the rest of reported performance. The Intelligence layer surfaces trends, risks, and investment opportunities quarterly.
The patent-pending methodology (GB2607796.6) covers the technical innovations that make this possible — the automated research pipeline, the jurisdiction-aware classification approach, and the synthesis of heterogeneous data into a coherent intangible asset view. The result is a platform that turns the measurement gap into a structured, continuous view of the intangible base of the business.
This is not a competing framework to financial reporting. It is a complementary layer that sits alongside the P&L and the balance sheet, doing for intangible capital what those instruments do for financial and physical capital. The companies that adopt it over the next few years will have a decision-making advantage that compounds with every AI investment they make.
The companion pieces in this series apply this framing to specific cases. Block's 40% workforce reduction examines the portfolio trade-offs most AI business cases don't model. The real ROI of AI isn't headcount reframes the whole investment case as capital formation rather than expense reduction.
Every technology wave produces its winners and losers, and the gap between them is always larger in hindsight than it looked in the moment. The winners of the AI economy will not be the companies that invested the most in AI. They will be the companies that measured the intangible capital their AI investments produced and allocated their subsequent investment accordingly. That is the discipline the measurement layer enables. It is the instrument the moment requires.
The companies building this discipline now will be the reference cases in three years. Two ways to start:
The measurement instrument the AI economy needs already exists. The question is which companies will put it in place before they need to explain to their board why the last £40M of AI investment didn't show up in enterprise value.
Measure your organization's productivity across multiple dimensions and benchmark against industry peers.
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 →
Research shows which intangible asset categories drive the greatest productivity gains in growth-stage companies and why balanced investment wins.
Read more →
A step-by-step guide to the Opagio Productivity Calculator — how to model your Gross Value Added, EBITDA, and Total Factor Productivity in real time. Learn what the numbers mean, how to interpret them, and why they matter for intangible asset investment decisions.
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.