The board meeting goes like this: the CTO presents the AI investment — new tools, new hires, new infrastructure. The CFO asks the obvious question: what is the return? The CTO offers a mix of anecdotes and activity metrics — tokens processed, models deployed, employees using AI daily. The board nods politely. No one in the room can connect the investment to a financial outcome, and the conversation moves on.
This scene plays out in boardrooms every quarter because traditional ROI frameworks were designed for investments in physical capital — machines, buildings, equipment — where the input and the output are both measurable and the time horizon is predictable. AI investment is fundamentally different. It creates intangible assets whose value compounds over time, generates optionality that may not be exercised for years, and produces gains that show up in revenue, efficiency, and competitive positioning simultaneously.
Measuring AI ROI with a traditional framework is like measuring a building's value by the cost of its bricks. The framework is not wrong — it is incomplete.
Traditional ROI calculations work on a simple premise: spend X, receive Y, calculate Y/X. This works for investments with direct, attributable returns — a new production line that increases output by a measurable amount, or a marketing campaign with trackable conversions. AI investment rarely fits this model for three reasons.
First, AI creates assets, not just savings. When a company invests in AI, it does not merely reduce costs. It builds proprietary datasets, trained models, organisational knowledge, and enhanced customer relationships. These are intangible assets that compound in value and generate returns across multiple business functions over extended time horizons. A traditional ROI calculation captures the cost reduction but misses the asset creation entirely.
Second, AI returns are non-linear. The value of AI investment follows a J-curve: initial deployment may actually reduce measured productivity while the organisation adapts. Returns then accelerate as complementary investments — in training, process redesign, data quality — compound with the technology investment. A 12-month ROI calculation captures the cost but misses the compounding.
Third, AI creates optionality. A trained model, a high-quality dataset, or an AI-literate engineering team creates strategic options that may not be exercised immediately. The option to enter a new market, to launch a new product, or to respond to a competitive threat faster has genuine value — but it does not appear in a traditional ROI calculation.
The fundamental error in most AI ROI frameworks is measuring only cost savings. AI creates intangible assets — technology capital, data assets, organisational knowledge — and the ROI framework must capture asset creation alongside efficiency gains. Without this, boards are making capital allocation decisions based on less than half the picture.
The framework that works for AI ROI has four distinct layers, each capturing a different type of value creation. Most boards measure only Layer 1 — efficiency gains — and miss three-quarters of the return.
This is the layer every board already understands: AI automates tasks, reduces costs, and increases throughput. It is the easiest to measure but often the least significant source of long-term value.
AI does not merely reduce costs — it drives revenue through better personalisation, faster responsiveness, improved product quality, and enhanced customer experience. Revenue lift is harder to attribute but more valuable than cost savings in most businesses.
This is the layer most frameworks miss entirely. Every AI deployment creates intangible assets: proprietary training data, fine-tuned models, institutional AI knowledge, enhanced brand capability. These assets have independent value — they can be licensed, sold, or leveraged for competitive advantage. The Opagio valuator assesses these assets across seven categories.
AI investment creates options — the option to enter new markets, launch new products, respond to competitive threats, or capitalise on emerging opportunities. Options have quantifiable value (real options theory provides the mathematics), but they require a different valuation approach than discounted cash flows.
The following table maps specific, measurable KPIs to each layer of the framework across five core business functions. This is the table your board presentation needs.
| Business Function | Layer 1: Efficiency | Layer 2: Revenue | Layer 3: Asset Creation | Layer 4: Optionality |
|---|---|---|---|---|
| Sales | Cost per qualified lead, sales cycle time | Conversion rate lift, average deal size | Customer propensity model, lead scoring dataset | Ability to enter new market segments |
| Operations | Process automation rate, error reduction | Throughput increase, quality premium | Process knowledge base, operational models | Capacity to scale without linear headcount |
| Product | Development velocity, defect rate | Feature adoption, NPS improvement | Proprietary algorithms, product IP | Platform extensibility for new products |
| Customer Service | Resolution time, cost per ticket | Retention rate, expansion revenue | Customer interaction dataset, service models | 24/7 multilingual capability |
| Finance | Reporting cycle time, audit cost | Forecast accuracy, working capital | Financial models, risk datasets | Real-time scenario analysis capability |
A portfolio company deployed AI-driven customer service and measured only Layer 1 (ticket resolution cost fell 35%). A deeper analysis revealed that Layer 2 (customer retention improved 12%, adding recurring revenue), Layer 3 (the trained customer service model became a licensable asset), and Layer 4 (the capability enabled expansion into three new languages without new hires) collectively represented four times the value of the efficiency gains alone.
There is a compounding effect that most ROI frameworks fail to capture: AI-created intangible assets do not merely add value — they multiply it. Each layer of the framework feeds the next.
Efficiency gains (Layer 1) generate data that improves AI models. Better models drive revenue lift (Layer 2). Revenue growth funds further AI investment, creating additional intangible assets (Layer 3). These assets expand strategic options (Layer 4), which — when exercised — create new efficiency gains and restart the cycle.
This is not a linear process. It is a flywheel. And the firms that understand this compound effect will deliver returns that linear ROI calculations systematically underestimate.
When AI creates a proprietary customer dataset, that asset improves prediction accuracy, which increases conversion rates, which generates more customer data, which further improves predictions. This is the intangible asset multiplier in action. Traditional ROI frameworks treat each improvement as independent. The 4-layer framework recognises that they compound — and that the measurement of intangible asset formation is essential to understanding the true return on AI investment.
Moving from metrics to narrative is where most AI ROI presentations fail. Boards do not need dashboards — they need a clear story that connects AI investment to outcomes they care about. Here is the structure that works.
Frame the AI investment as a capital allocation decision, not a technology decision. "We invested X in AI to build Y intangible assets that will generate Z in returns over N years." This is the language boards speak.
Show the board that you measure beyond cost savings. Present KPIs from each layer using the table above. The breadth of measurement demonstrates management rigour.
Name the intangible assets AI has created or enhanced: proprietary datasets, trained models, institutional knowledge, enhanced customer relationships. Use the Opagio framework to assign indicative values. This transforms AI from a cost line into a capital investment.
Demonstrate how current AI assets will generate increasing returns over time. Use the flywheel narrative: efficiency gains create data, data improves models, models drive revenue, revenue funds further investment.
Boards respond to risk as much as opportunity. Frame the alternative: competitors that build AI-driven intangible assets now will have a compounding advantage that becomes increasingly expensive to close.
Three errors appear repeatedly in AI ROI presentations. Avoiding them substantially improves board engagement and capital allocation decisions.
Measuring only cost savings. This is the most common and most damaging error. It reduces AI to an efficiency tool and ignores the three layers where the majority of value is created. Boards that see only cost savings will — rationally — question whether the investment is worthwhile relative to simpler automation.
Ignoring organisational learning. The knowledge that a team accumulates while deploying and using AI — how to structure data, how to evaluate model outputs, how to design AI-augmented workflows — is itself a valuable intangible asset. Companies that measure only the model's output miss the institutional capability it builds.
Using AI metrics instead of business metrics. Boards do not care about model accuracy, inference latency, or token throughput. They care about revenue, margin, customer retention, and competitive positioning. Every AI metric in a board presentation must be translated into a business metric. If the translation is not possible, the metric should not be in the presentation.
From my experience at IG Group, the most effective technology ROI presentations always started with the business outcome and worked backward to the technology. The board never asked "what model architecture did you use?" — they asked "how did this affect customer acquisition cost?" Frame every AI discussion in terms the board already uses to evaluate the business.
At IG Group, we invested heavily in technology across 15 years — building platforms, mobile applications, trading engines, and customer-facing innovations that drove the company from a valuation of roughly three hundred million pounds to nearly three billion. The City frequently questioned why the technology cost base was so high.
The answer was always the same: technology investment was not a cost — it was asset creation. Every platform we built, every mobile app we launched, every customer interaction we automated created intangible assets that generated revenue for years. Our first mobile trading platform, launched in 2004, was still generating a significant share of company revenue more than a decade later. The ROI calculation that captured only the initial development cost would have dramatically underestimated the return.
This experience directly informed the Opagio growth platform. The gap between what technology investment costs and what it creates is precisely the gap that traditional ROI frameworks — and traditional accounting standards — fail to capture. The 4-layer framework exists to close that gap.
Ivan Gowan is Founder and CEO of Opagio. He spent 15 years as a senior technology leader at IG Group (LSE: IGG), overseeing engineering growth from 4 to 250 during the company's rise from £300m to £2.7bn. He built IG's first online and mobile trading platforms, launched the world's first Apple Watch trading app, and holds an MSc from Edinburgh with neural networks research (2001). Learn more about the Opagio team.
Board oversight of AI investment is becoming as critical as financial controls. Here is the 10-point accountability checklist that separates effective board governance from box-ticking.
Read more →
Global AI investment has surpassed $500 billion annually, yet total factor productivity growth remains stubbornly flat. This is Solow's paradox reborn: you can see AI everywhere except in the productivity statistics. History tells us why — and what needs to change.
Read more →
Individual developers using AI copilots write code 30-55% faster. Individual analysts produce reports in half the time. Yet organisational productivity remains flat. The gap between task-level AI gains and enterprise-level outcomes reveals a measurement failure — and an intangible asset opportunity — that most firms are missing entirely.
Read more →Get the latest insights on intangible asset growth and productivity delivered to your inbox.
Book a free consultation to see how the Opagio Growth Platform can help your business.