The AI ROI Dilemma: How to Stop Gambling and Start Investing Wisely in Technology

Investing in artificial intelligence is no longer an option—it has become a competitive imperative. Business leaders across Europe are allocating six-figure and larger budgets, with 78% of organizations planning to increase their spending on this technology. However, behind this enthusiasm lies a harsh truth confirmed by hard data: the vast majority, between 70% and even 98% of AI initiatives, never achieve their intended business goals.

For boards and CFOs, this means facing a fundamental dilemma: how to reconcile the need for innovation with the risk of burning through millions on a project with an uncertain return on investment (ROI)? As Deloitte points out, for many companies, a measurable ROI remains a "big question mark," which is the main barrier to decision-making.

In this analysis, based on data from the world's leading research firms, we prove that the problem isn't the technology. It's the strategy.

Why Traditional ROI Fails in the World of AI

Trying to measure the value of an AI project with a standard spreadsheet is like measuring an engine's temperature with a barometer—we're using the wrong tool. The value generated by AI is not static; it evolves, compounds over time, and often eludes simple calculations for several reasons:

• Dominance of Intangible Benefits: A significant portion of AI's value lies in "soft" benefits, such as improved decision quality, increased customer satisfaction, and building a competitive advantage. These elements, while crucial, are extremely difficult to quantify in short-term financial models.
• Long Time-to-Value: The full value of AI often materializes over the long term, requiring time for model training, iteration, and organizational adaptation.
• Unpredictable and Escalating Costs: Gartner warns that miscalculating the costs of scaling generative AI can lead to errors of 500% to 1000%. More than half of organizations abandon projects precisely because of unforeseen costs.

The Anatomy of Failure – Where Does the Real Problem Lie?

The biggest mistake is treating AI implementation as a purely technical challenge. The research from Boston Consulting Group is unforgiving: about 70% of barriers are organizational and process-related, technology accounts for 20%, and the algorithms themselves only 10% of the problems. Market leaders instinctively apply the "10-20-70 rule," allocating 70% of their resources to human and process transformation. As McKinsey & Company proves, the factor with the greatest impact on operating profit (EBIT) generated by AI is "workflow redesign."

How Smart Leaders De-Risk AI Investments

Instead of making a single, monolithic bet, market leaders transform the investment into a series of smaller, evidence-based decisions. Key strategies include:

1. Phased Implementation (PoC -> MVP -> Scale): An iterative model allows for gradual learning and validation of assumptions, starting with a limited "Proof of Concept" to provide hard data for the business case.
2. Holistic Evaluation Frameworks: It's necessary to go beyond simple ROI and implement advanced frameworks like Forrester's Total Economic Impact™ (TEI), which considers benefits, costs, flexibility, and risk.
3. Building Lasting Value Over Technology: One of the biggest hidden risks is the lack of people capable of operating and developing new technology. At co.brick, we understand that the true value of AI is unlocked when an organization builds its own internal competencies. Therefore, instead of offering clients a "black box," our approach is based on building their strategic independence. In models like Build-Operate-Transfer (BOT), we not only deliver a solution but, most importantly, build a dedicated, autonomous team from the ground up, only to transfer it to the client's ownership after an agreed-upon time. We believe this is the most powerful de-risking strategy, ensuring that an investment in technology transforms into a durable, internal company asset, not a dependency on a single vendor.

From Theory to Practice – How Market Leaders Measure Success

The best companies don't focus on abstractions but on hard metrics. Here are some examples from the report:

• Siemens, by optimizing production planning with AI, achieved a 15% reduction in production time and a 12% decrease in costs.
• American Express, by implementing an AI chatbot, saw a 25% reduction in customer service operational costs and a 10% increase in satisfaction (CSAT).
• General Mills, thanks to AI in logistics, has saved over $20 million in transportation costs since 2024.

Conclusion: Change the Question from "What's the ROI?" to "How Do We Build Value?"

Success in the AI era doesn't depend on the size of the budget, but on strategic maturity. Instead of asking, "What will be the return on this tool?" leaders must ask, "How can we redesign our business to fully leverage the potential of this technology, and who can help us do it?" This shift in perspective from buying technology to building a strategic partnership is the key to turning a risky bet into certain, long-term value.

Selected Sources:

• McKinsey, "The state of AI: How organizations are rewiring to capture value"
• Boston Consulting Group (BCG), "AI Adoption in 2024: 74% of Companies Struggle to Achieve and Scale Value"
• Gartner, "Generative AI: What Is It, Tools, Models, Applications and Use Cases"
• Forrester, "Total Economic Impact Methodology"
• Deloitte, "State of Generative AI in the Enterprise 2024"
• Harvard Law School, "Artificial Intelligence: An engagement guide"

Enjoyed this article? Don't miss out! Join our LinkedIn community and subscribe to our newsletter to stay updated on the latest business and innovation news. Be part of the conversation and stay ahead of the curve.

Business Development Manager

Tomasz Ziaja