--- title: "AI Business Cases" date: 2026-07-12 description: Business cases built for predictable payback misread AI, whose value arrives in parallel, late, and elsewhere; Boards need different instruments. author: Mario Thomas canonical: https://mariothomas.com/briefings/ai-business-cases/ --- ## Start here Two short reads before the deeper work: what an AI investment case has to do, and the order to take this briefing in. ### Why AI Breaks the Business Case In 2015 I built AWS's first comprehensive cloud business case tool, and the discipline it embodied served Boards well because cloud adoption followed a sequential, predictable path that TCO, ROI, and NPV could capture. AI does not behave that way. Adoption runs in parallel at different speeds across the organisation, costs shift as models and hardware evolve, and value routinely surfaces in functions other than the one that funded the work. McKinsey's State of AI research (2025) found more than **80%** of organisations had seen no tangible enterprise-level EBIT impact from generative AI, and I would argue the instruments Boards use to evaluate the spend are part of why. The pattern I see most often is one evaluation framework and one ROI threshold applied to every AI proposal. That misallocates capital in both directions, smothering quick wins with excessive rigour while waving through high-risk initiatives with too little scrutiny. It also treats the foundational work, the data cleaned, the governance built, the skills developed, as project overhead rather than as capability that compounds across every subsequent initiative. The position these articles take is that financial discipline is not the casualty; the instruments are the problem. Value should be assessed across the five pillars of the Well-Advised principle rather than cost reduction alone, tracked through leading, lagging, and predictive indicators rather than lagging measures only, and held as a portfolio of quick wins, capability builds, and long-horizon initiatives rather than a queue of isolated projects. And the largest returns, the second- and third-order effects, tend to arrive later and elsewhere than the original case anticipated. Read this briefing and you should be able to make two judgements about the next AI proposal that reaches the Board: whether the case in front of you deserves approval, and whether the Board's evaluation machinery is capable of seeing the value, and the cost, at stake. ### Read the Core Sequence First The heart of this briefing is the Rethinking Business Cases in the Age of AI series, written in order, and it rewards being read that way. Start with What Boards Need to Know, which makes the case that conventional models cannot see AI's value creation patterns. Creating the Foundation then sets out the five building blocks of a better evaluation: strategic purpose, the value spectrum, the true investment profile, readiness, and scaling potential. The sequence then turns practical. Finding High-Value AI Opportunities gives you a systematic way to surface the initiatives worth evaluating at all, Building Your AI Business Case assembles the case itself step by step, and Securing Buy-In from the Board addresses the moment the case meets the Boardroom, including the six areas of concern directors consistently raise. The sequence sits under [Core reading](#core-reading); the pieces alongside, on cost and measurement, sit under [Further reading](#further-reading). The further reading goes deeper on cost, and on measurement and its limits. Read The Balancing Item alongside Creating the Foundation: it prices the oversight labour the True Investment Profile warned about. Measuring AI Value introduces the Well-Advised principle the whole series leans on, AI's Hidden ROI extends measurement to second- and third-order effects, and The Appreciating Ledger takes the argument to the CFO and the audit committee. Behind those sit Maximum Fidelity, which supplies the four indicator types a Board measures with, and Transforming the Board, where decision analytics becomes the Board's own instrument; Not Everything Needs AI asks the questions that come before any decision to build. Read Avoiding the Business Case Trap last: it is the corrective that no accumulation of well-crafted business cases, however rigorous, amounts to a strategy. The mechanisms beneath the thinking, including the AI Stages of Adoption model and Well-Advised, are gathered under [Remake](#remake-assets), and [the questions](#faqs) give you a fast check on whether your own organisation's approach would survive the scrutiny this briefing recommends. ## Core reading The core sequence, from why conventional models fail to a case the Board can approve, with deeper pieces on cost and measurement alongside. 1. [Rethinking Business Cases in the Age of AI: What Boards Need to Know](https://mariothomas.com/blog/ai-business-case-new-thinking/) (11 minute read, 22 April 2025): Traditional business case methods assume sequential adoption. AI runs in parallel across maturity stages at once, and the valuation has to change with it. 2. [Rethinking Business Cases in the Age of AI: Creating the Foundation](https://mariothomas.com/blog/ai-business-case-foundation/) (14 minute read, 28 April 2025): AI's parallel, multi-speed adoption breaks the sequential logic of traditional business cases. These are the building blocks an evaluation needs before any number is written. 3. [Rethinking Business Cases in the Age of AI: Finding High-Value AI Opportunities](https://mariothomas.com/blog/ai-business-case-finding-opportunities/) (15 minute read, 4 May 2025): The highest-value AI opportunities rarely sit in the fashionable applications. They sit in the invisible, strategically significant processes a structured evaluation surfaces. 4. [Rethinking Business Cases in the Age of AI: Building Your AI Business Case](https://mariothomas.com/blog/ai-business-case-step-by-step/) (18 minute read, 12 May 2025): A disciplined AI business case balances financial rigour with AI's unusual value patterns: the step-by-step construction, from opportunity to a case a Board can sign. 5. [Rethinking Business Cases in the Age of AI: and Securing Buy-In from the Board](https://mariothomas.com/blog/ai-business-case-secure-buy-in/) (16 minute read, 19 May 2025): A well-built AI business case can still fail at the Board. When 88% of pilots never reach production, approval must carry a path to scale. ## Further reading - [The Balancing Item: The AI Oversight Cost Your Business Case Never Priced](https://mariothomas.com/blog/unpriced-cost-ai-oversight/) (10 minute read, 26 July 2026): Every AI business case counts the hours saved. Almost none counts the oversight hours added, and people are silently absorbing the difference. Podcast edition: 13 minute listen. - [Measuring AI value: A strategic framework for Boards and business leaders](https://mariothomas.com/blog/measuring-ai-roi/) (12 minute read, 13 January 2025): Measuring AI value needs more than total cost of ownership. A strategic framework for Boards, built on what the cloud business case taught me. - [AI’s Hidden ROI: Measuring Second and Third-Order Effects for Board Decisions](https://mariothomas.com/blog/ai-hidden-roi-cascade/) (11 minute read, 10 August 2025): AI's largest returns arrive late, as second- and third-order effects on capability and business model. Boards need leading and predictive indicators to see them. - [The Appreciating Ledger: When AI Capital Outgrows the CFO's Rulebook](https://mariothomas.com/blog/cfo-appreciating-ledger/) (11 minute read, 19 April 2026): AI capital appreciates, accumulates, and crosses functions. The four indicator types let CFOs see what conventional project-ROI models structurally cannot. Podcast edition: 13 minute listen. - [From AI Pilots and Projects to AI Strategy: Avoiding the Business Case Trap](https://mariothomas.com/blog/ai-strategy-business-case-trap/) (10 minute read, 5 October 2025): Approving AI business cases at record pace is not a strategy: 92% are investing more, 1% have reached maturity. The gap is coherence. - [Maximum Fidelity: How Four Indicator Types Strengthen Board Decisions](https://mariothomas.com/blog/maximum-fidelity-four-indicators/) (13 minute read, 12 April 2026): Four indicator types give boards progressively higher decision fidelity: lagging, leading, predictive, and reasoned. Together they represent the most accountable governance instrument available. Podcast edition: 15 minute listen. - [Transforming the Board: Using Decision Analytics for Strategic Advantage](https://mariothomas.com/blog/board-ai-decision-analytics/) (13 minute read, 23 March 2025): Decision analytics moves the Board from backward-looking metrics to predictive indicators that model possible futures, and changes how directors exercise judgement. - [Not Everything Needs AI: The Questions That Come Before the Decision](https://mariothomas.com/blog/not-everything-needs-ai/) (8 minute read, 12 July 2026): Boards keep being told to remake the business around AI. The real question is not which tool, but whether the work needs doing at all. Podcast edition: 8 minute listen. ## Remake The mechanisms beneath the thinking: the model, diagnostic, methodology, and principle from the Remake Library that turn this briefing into apparatus a Board can use. - **Model: Maximum Fidelity**. The evidence discipline of grading every reading a Board relies on by one of four indicator types: lagging indicators of past outcomes, leading indicators of early signals, predictive indicators of future value, and reasoned indicators that prove what must hold true. [Remake Library](https://mariothomas.com/remake/library/#maximum-fidelity) - **Diagnostic: AI Initiative Rubric**. A pilot evaluation tool that scores candidate initiatives across the five Well-Advised value priorities and Five Pillars capability building, recommending whether to prioritise, defer, pipeline, or develop them further. [Remake Library](https://mariothomas.com/remake/library/ai-initiative-rubric/) - **Methodology: AI Business Case**. The integrated decision framework that crystallises across an ADAPT engagement rather than at a single stage: strategic alignment established at Align, cost and readiness evidenced at Diagnose, value shaped at Advise, and execution designed at Plan. [Remake Library](https://mariothomas.com/remake/library/#ai-business-case) - **Principle: Well-Advised**. The framework of five strategic priorities, Innovation, Customer Value, Operational Excellence, Responsible Transformation, and Revenue, used to ensure AI investments create balanced value rather than narrow cost reduction. [Remake Library](https://mariothomas.com/remake/library/well-advised/) ## Questions Questions directors ask me about AI investment, answered from the work gathered in this briefing. ### Why can't we evaluate AI investments the way we evaluated cloud or previous technology? Cloud adoption followed a sequential path with predictable cost categories, which is why TCO and ROI models served it well. AI adoption runs in parallel across the organisation, with initiatives at different maturity stages, on different timelines, and creating value in functions other than the one that funded them. A uniform threshold applied across that landscape misallocates capital in both directions. I set out the full argument in [Rethinking Business Cases in the Age of AI: What Boards Need to Know](/blog/ai-business-case-new-thinking/). ### Our pilots succeed but nothing reaches production. What are we getting wrong? IDC research, reported by CIO.com in 2025, found that 88% of AI pilots never reach production, and in my experience the cause is rarely the technology. Cases are approved without predefined scaling triggers, handover processes, or governance designed to evolve from experiment to production, so pilots become curiosities. [Securing Buy-In from the Board](/blog/ai-business-case-secure-buy-in/) sets out what to build into the approval itself. ### How do we measure return on an investment whose largest benefits haven't arrived yet? By widening the instrument set. Lagging indicators confirm value after the fact; leading and predictive indicators signal it early enough to govern with, and the [Well-Advised principle](/blog/measuring-ai-roi/) organises them across the five dimensions Boards consistently care about. The largest returns are typically [second- and third-order effects](/blog/ai-hidden-roi-cascade/) that mature over 12 to 24 months or longer, and Boards that measure only first-order gains risk abandoning initiatives whose real value lies ahead. ### Should every AI initiative have to clear the same ROI hurdle? No. A healthy AI portfolio holds three different kinds of investment: quick wins that fund momentum, capability builds whose value is the initiatives they enable, and long-horizon initiatives judged on option value and strategic position. Imposing a single hurdle rate collapses that distribution into one category and starves the foundations. I develop the portfolio approach in [Finding High-Value AI Opportunities](/blog/ai-business-case-finding-opportunities/) and the finance-function version in [The Appreciating Ledger](/blog/cfo-appreciating-ledger/). ### Do good AI business cases add up to an AI strategy? They do not, and this is the trap. A business case is a tool of isolation: it evaluates one initiative against narrow criteria and cannot ask whether the pieces reinforce or undermine each other. Organisations that govern AI project by project tend to accumulate incompatible data models, duplicated governance, and pilots that never compound. [Avoiding the Business Case Trap](/blog/ai-strategy-business-case-trap/) explains why the Board's role is orchestrating systematic capability, not gatekeeping proposals. ### What costs do our AI business cases routinely leave out? Data preparation first. In my experience initial budgets allocate 15 to 20% to it, and organisations end up spending 50 to 65% of project cost cleaning, structuring, and governing data. Then the recurring lines: model monitoring and retraining, governance overhead, specialised talent, and parallel running of old and new systems during transition. The newest omission is oversight labour, the reviewing and correcting AI outputs need before anyone can rely on them; BCG Henderson Institute research in Harvard Business Review (2026) found 14% of workers using AI reporting the strain. The [True Investment Profile](/blog/ai-business-case-foundation/) sets out the cost categories, and [The Balancing Item](/blog/unpriced-cost-ai-oversight/) prices the one almost every case misses. ### Should we treat AI capability as an asset rather than an expense? On the internal scorecard, yes. The capability AI investment builds (the cleaned data, governance, prompt libraries, and trained workforce) appreciates through use and is reused by initiatives other than the one that paid for it, which no depreciation schedule records. IAS 38 already sets four tests for an intangible asset, identifiability, control, measurability, and future economic benefit, and modern AI systems meet them; the rules permit the treatment, practice has not caught up. The practical move is to measure it on the management report while the statutory accounts still expense it, and to brief the audit committee early. [The Appreciating Ledger](/blog/cfo-appreciating-ledger/) makes the case to the CFO. ### How do we find the AI opportunities worth a business case? Start with processes rather than products; the fashionable applications are rarely where the value sits. I score each candidate process on decision complexity, information volume, process variability, cognitive repetition, and knowledge continuity risk, and look hardest at those that score high on several. The shortlist then passes through the Well-Advised principle, which favours initiatives delivering across several pillars over a spike in one, and a feasibility check against the Five Pillars diagnostic. One question comes before all of that: whether the work needs doing at all. [Finding High-Value AI Opportunities](/blog/ai-business-case-finding-opportunities/) sets out the method, and [Not Everything Needs AI](/blog/not-everything-needs-ai/) the question that precedes it. ## References The external research these articles draw on, for directors who want the evidence at first hand. - **McKinsey & Company** (5 November 2025): [The State of AI](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai). The recurring global survey behind the finding that most organisations see no enterprise-level EBIT impact from generative AI. - **McKinsey & Company** (28 January 2025): [Superagency in the workplace: Empowering people to unlock AI’s full potential](https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/superagency-in-the-workplace-empowering-people-to-unlock-ais-full-potential-at-work). Evidence that organisations scaling AI across functions capture up to four times more value than those confined to pilots. - **Bain & Company** (12 April 2026): [CFOs Funded the AI Revolution. Now They’re Joining It.](https://www.bain.com/insights/cfos-funded-ai-revolution-now-they-are-joining-it/). The April 2026 CFO survey showing satisfaction with AI outcomes rises with the adoption curve, from 31% overall to over 60% at the top. - **PwC** (13 April 2026): [Three-quarters of AI’s economic gains are being captured by just 20% of companies – with the leading companies focused on growth, not just productivity](https://www.pwc.com/gx/en/news-room/press-releases/2026/pwc-2026-ai-performance-study.html). The 2026 AI Performance Study of 1,217 executives, finding 74% of AI's economic value captured by 20% of organisations. - **BCG** (30 September 2025): [AI Leaders Outpace Laggards with Double the Revenue Growth and 40% More Cost Savings](https://www.bcg.com/press/30september2025-ai-leaders-outpace-laggards-revenue-growth-cost-savings). BCG's September 2025 release on The Widening AI Value Gap, the survey of 1,250 executives behind the 5% of future-built organisations outpacing the rest. - **BCG** (4 June 2025): [How to Get ROI from AI in the Finance Function](https://www.bcg.com/publications/2025/how-finance-leaders-can-get-roi-from-ai). The March 2025 survey of 280 finance executives behind the 10% median reported AI ROI, and what the outperformers share. - **Gartner** (25 June 2025): [Gartner: Over 40% of Agentic AI Projects Will Be Canceled by End 2027](https://www.gartner.com/en/newsroom/press-releases/2025-06-25-gartner-predicts-over-40-percent-of-agentic-ai-projects-will-be-canceled-by-end-of-2027). Gartner's 2025 prediction that over 40% of agentic AI projects will be cancelled by the end of 2027, citing escalating costs, unclear business value, and inadequate risk controls. - **S&P Global** (30 May 2025): [Generative AI experiences rapid adoption, but with mixed outcomes – Highlights from VotE: AI & Machine Learning](https://www.spglobal.com/market-intelligence/en/news-insights/research/ai-experiences-rapid-adoption-but-with-mixed-outcomes-highlights-from-vote-ai-machine-learning). The 2025 analysis finding 42% of businesses scrapped most of their AI initiatives, up from 17% a year earlier. - **RAND Corporation** (13 August 2024): [The Root Causes of Failure for Artificial Intelligence Projects and How They Can Succeed](https://www.rand.org/pubs/research_reports/RRA2680-1.html). The root-cause research behind the widely cited 70-85% AI project failure estimates. - **Brookings Institution** (15 January 2026): [Counting AI: A blueprint to integrate AI investment and use data into US national statistics](https://www.brookings.edu/articles/counting-ai-a-blueprint-to-integrate-ai-investment-and-use-data-into-us-national-statistics/). The January 2026 blueprint on counting AI as investment rather than operating cost, the same correction the ledger needs inside organisations. - **Harvard Business Review** (1 March 2026): [When Using AI Leads to “Brain Fry”](https://hbr.org/2026/03/when-using-ai-leads-to-brain-fry). BCG Henderson Institute research in HBR: a survey of 1,488 full-time US workers identifying mental fatigue that attaches to heavy AI oversight loads, with replacement-pattern AI use associated with lower burnout. - **IDC** (25 March 2025): [88% of AI pilots fail to reach production — but that’s not all on IT](https://www.cio.com/article/3850763/88-of-ai-pilots-fail-to-reach-production-but-thats-not-all-on-it.html). IDC's finding, reported by CIO.com in 2025, that 88% of AI pilots never reach production, and the case that the causes sit well beyond IT. ## The ideas beneath this briefing Ideas I've named and matured writing about AI Business Cases: what each one means, and where it started. - **True Investment Profile**: A fuller accounting of AI cost that captures data preparation, retraining cycles and governance overhead, which conventional ledgers systematically undercount. [Read more](https://mariothomas.com/blog/cfo-appreciating-ledger/) - **Business Case Trap**: The misconception that accumulating individual AI business cases, each justified in isolation, constitutes AI strategy, creating fragmentation, incompatible governance and pilots that never cohere into competitive advantage. [Read more](https://mariothomas.com/blog/ai-strategy-business-case-trap/) All concepts: https://mariothomas.com/glossary/concepts/ ## More Board Briefings More complete resources on AI and emerging technology for the Boards that need the full picture. - [AI Strategy](https://mariothomas.com/briefings/ai-strategy/): Approving good AI projects is not a strategy, and the Board's move is from accumulating pilots to a strategy it owns. - [Operating AI](https://mariothomas.com/briefings/operating-ai/): Most AI pilots never reach production; the AI Centre of Excellence is the operating capability that turns scattered experiments into governed, scaled adoption. - [AI & the Board](https://mariothomas.com/briefings/ai-and-the-board/): AI changes how every Board duty is discharged, from director to Company Secretary, and moves none of the accountability. - [Data Strategy](https://mariothomas.com/briefings/data-strategy/): Every AI system runs on data the balance sheet cannot see, so Boards should govern, structure, and monetise it as the asset it is.