University of Kwazulu-Natal

Rethinking Ai Roi Why Traditional Investment Models No Linkedin

Bonisiwe Shabane
-
rethinking ai roi why traditional investment models no linkedin

In today’s rapidly evolving technological landscape, IT has evolved from a traditional operational backbone to a strategic business partner. This shift is fueled by accelerating digitalization and automation, including the rise of AI use cases that deliver value well beyond the IT function. Leading companies who have successfully made this transition are already capturing results: Visa leveraged generative AI to enhance fraud detection, preventing $40 billion in fraud between October 2022 and September 2023, twice as much... Yet, as organizations pour resources into similar AI initiatives, they confront a pressing dilemma: how to accurately measure AI ROI when traditional investments overlook AI’s distinctive value dynamics. AI investments require a redefined ROI paradigm, one that moves beyond one-size-fits-all models, to account for their non-linear returns, attribution complexities, and the contextual forces that shape their impact. AI investments diverge significantly from conventional investment in how they generate and demonstrate value.

While traditional projects typically follow predictable implementation curves with clear operational impacts, AI initiatives offer a more complex value proposition: benefits evolve over time, are often indirect and resist clear isolation or attribution to... This fundamental difference requires organizations to reconsider how they approach ROI calculation and justification for AI spend. Let’s first explore why AI returns tend to accelerate over time rather than follow a steady path, and then examine attribution complexities, one of the most overlooked challenges in assessing AI ROI. In today’s rapidly evolving technological landscape, IT has evolved from a traditional operational backbone to a strategic business partner. This shift is fueled by accelerating digitalization and automation, including the rise of AI use cases that deliver value well beyond the IT function. Leading companies who have successfully made this transition are already capturing results: Visa leveraged generative AI to enhance fraud detection, preventing $40 billion in fraud between October 2022 and September 2023, twice as much...

Yet, as organizations pour resources into similar AI initiatives, they confront a pressing dilemma: how to accurately measure AI ROI when traditional investments overlook AI’s distinctive value dynamics. AI investments require a redefined ROI paradigm, one that moves beyond one-size-fits-all models, to account for their non-linear returns, attribution complexities, and the contextual forces that shape their impact. AI investments diverge significantly from conventional investment in how they generate and demonstrate value. While traditional projects typically follow predictable implementation curves with clear operational impacts, AI initiatives offer a more complex value proposition: benefits evolve over time, are often indirect and resist clear isolation or attribution to... This fundamental difference requires organizations to reconsider how they approach ROI calculation and justification for AI spend. Let’s first explore why AI returns tend to accelerate over time rather than follow a steady path, and then examine attribution complexities, one of the most overlooked challenges in assessing AI ROI.

“What we’re seeing is that accessibility-driven design often solves a broader problem. It’s not charity. It’s engineering.” Telangana has attracted over 75 greenfield GCCs in 2025, compared with 40-plus in Karnataka. “Only 30% of software engineering happens on the laptop. The real 70% starts after you commit the code,” says Jyoti

From groundwater and slopes to carbon sinks, tools like CatBoost are enabling Indian scientists to extract insights and drive sustainability. With capacity expected to more than double this decade, the industry is investing in training as graduates struggle to meet Yet despite this momentum, most respondents reported achieving satisfactory ROI on a typical AI use case within two to four years. This is significantly longer than the typical payback period of seven to 12 months expected for technology investments.1 Only six per cent reported payback in under a year, and even among the most successful... Deloitte’s 2025 survey of 1,854 executives across Europe and the Middle East, supported by 24 in-depth interviews, shows that momentum is building. In ten percent of organizations, the CEO is the primary leader of the AI agenda.

Increasingly organizations view AI as a strategic imperative, not just a technology upgrade – especially as agentic AI begins to reshape assumptions about how businesses will operate in the future. To capture value, leading enterprises are shifting towards CEO-led, organization-wide prioritization of AI. They are also becoming more selective in their choice of use cases and are building structured programs to drive the more profound organizational change needed to scale AI across the business. Generative AI is already delivering measurable productivity gains. Agentic AI involves greater complexity but holds the potential for end-to-end process redesign. However, embedding AI into the fabric of an organization is not a simple upgrade; it is akin to the transition from steam to electricity.

When factories switched from steam power, they had to reconfigure their production lines, redesign workflows, invest in new infrastructure and reskill their workforce. The full benefits only emerged once organizations fundamentally changed how they operated. The same is true for AI. It demands significant planning, long-term investment and often deep organizational change. Over time, AI will become embedded into core operations, reshaping how businesses create value. As AI becomes embedded in the fabric of businesses, a familiar challenge is taking new shape: how to measure return on investment.

Traditional ROI models, geared towards immediate, quantifiable gains, often fall short in capturing the full picture of AI’s value. To succeed, organisations must adopt a grounded, forward-looking, and holistic approach to defining AI ROI – one that is essential not just for short-term validation, but for sustained success and transformational impact. Thinking about AI investments? Here are five things to keep in mind when assessing ROI. A common mistake when adopting AI is using traditional ROI models that expect quick, obvious wins like cutting costs or boosting revenue right away. But AI works differently.

Its real value often shows up slowly, through better decision-making, greater agility, and preparing the organisation to compete long-term. AI projects need big upfront investments in things like improving data quality, upgrading infrastructure, and managing change. These costs are clear from the start, while the bigger benefits, like smarter predictions, faster processes, and a stronger competitive edge, usually take years to really pay off and aren’t easy to measure the... When companies invest in artificial intelligence, the first question often asked is: “What’s the ROI?” It's a fair question - but often the wrong one, at least in the way it's traditionally framed. In the early 2000s, no one could give you an ROI calculation for building a website.

But they could tell you: if you don’t, you’ll be irrelevant. And I think we’re seeing the same inflection point now. The mistake companies make is trying to apply the same ROI model they used for upgrading a server or rolling out a CRM. AI is not a one-time purchase or a bolt-on tool. It’s a foundational capability shift. It changes how you think, decide, and evolve (who does what, in what order, and with what automation handoffs), not just what you can automate.

Classic ROI models focus on quantifiable, short-term outcomes: Having sat on two panels on this topic and participated in numerous client discussions, I've seen firsthand a fundamental shift in how organizations approach technology investment decisions. We need a different approach to evaluating these investments - one that accounts for their unique characteristics and strategic importance. For decades, technology investments followed predictable patterns. When replacing a block of PCs or evaluating infrastructure upgrades, organizations could rely on straightforward calculations with clearly defined parameters. You could identify direct cost savings and project revenue impacts and calculate a reliable ROI within a specific timeframe.

Emerging technologies operate on entirely different principles. The investment is primarily geared toward understanding the technology and how it might impact your business rather than delivering immediate, quantifiable returns. Consider AI implementations. Most organizations begin with proof-of-concept initiatives aimed not at immediate cost reduction but at exploring capabilities and potential applications. The primary value comes from learning and positioning rather than immediate operational improvements. This exploration-focused approach doesn't fit neatly into traditional ROI templates.

There's rarely a definitive calculation where you can predict saving X dollars by implementing Y technology. Businesses have long measured AI investments through cost savings, efficiency, and automation. But does that approach still hold when AI is no longer just a tool but the foundation of enterprise strategy? AI is no longer a standalone investment — it’s embedded in how companies operate, compete, and grow. Measuring its success through short-term financial metrics overlooks the bigger picture. Instead of asking, “What’s the ROI of AI?” the real question is: “How do we ensure AI is fully embedded, optimized, and driving sustained competitive advantage across the business?”

AI, including GenAI, is still in its adoption curve — executives often require ROI justification before scaling investments. But will we even need to measure AI ROI in the future? AI follows the same trajectory. Initially, companies measured its financial impact, but as it becomes a core driver of enterprise agility, risk management, and competitive advantage, AI is shifting from an IT initiative to an operational necessity. Yet, in today’s reality, the pressure to justify AI’s financial return remains high. According to the Dataiku trends report, 85% of data, analytics, and IT leaders are under pressure from the C-suite to quantify ROI from GenAi.

And while 72% report positive returns from their GenAI projects, measurement remains a challenge. When I was working as a portfolio manager at top investment firms, I’d come in each morning to hundreds of research reports waiting for me. Despite having a team of brilliant analysts, we were constantly behind‌ — ‌not because we weren’t smart enough, but because the sheer volume of information was impossible for humans to process effectively. What if we miss critical insights buried in pages of research, or by the time we’d synthesized everything, market conditions had already shifted? The financial industry produces no shortage of investment research. Research firms generate hundreds of reports per day, and analysts and financial advisors are easily buried in this avalanche of information.

Somehow, they’re expected to sift through it all to find critical insights — while managing their other responsibilities. Here’s the thing — you can’t just throw more people at this problem. I’ve seen firms try to hire their way out of information overload, and it doesn’t work. The volume of investment research will always increase faster than any team can handle. The answer is agentic AI, with purpose-built systems designed for expert, accurate analysis. When I first encountered generative AI during my career in financial services, I had one of those moments where everything clicked.

This technology could help me find market-driving insights and separate signal from noise in ways I’d never imagined. What I realized is that AI isn’t limited by volume the way humans are. At WRITER, I work with analysts and financial advisors who rely on data from research reports, news articles, company 10-K filings, and alternative sources‌ — ‌satellite imagery, hiring trends, social media sentiment, supply chain... These sources provide so much more nuance and depth than numbers alone, but processing them manually? That’s where teams get stuck.

People Also Search

In Today’s Rapidly Evolving Technological Landscape, IT Has Evolved From

In today’s rapidly evolving technological landscape, IT has evolved from a traditional operational backbone to a strategic business partner. This shift is fueled by accelerating digitalization and automation, including the rise of AI use cases that deliver value well beyond the IT function. Leading companies who have successfully made this transition are already capturing results: Visa leveraged g...

While Traditional Projects Typically Follow Predictable Implementation Curves With Clear

While traditional projects typically follow predictable implementation curves with clear operational impacts, AI initiatives offer a more complex value proposition: benefits evolve over time, are often indirect and resist clear isolation or attribution to... This fundamental difference requires organizations to reconsider how they approach ROI calculation and justification for AI spend. Let’s firs...

Yet, As Organizations Pour Resources Into Similar AI Initiatives, They

Yet, as organizations pour resources into similar AI initiatives, they confront a pressing dilemma: how to accurately measure AI ROI when traditional investments overlook AI’s distinctive value dynamics. AI investments require a redefined ROI paradigm, one that moves beyond one-size-fits-all models, to account for their non-linear returns, attribution complexities, and the contextual forces that s...

“What We’re Seeing Is That Accessibility-driven Design Often Solves A

“What we’re seeing is that accessibility-driven design often solves a broader problem. It’s not charity. It’s engineering.” Telangana has attracted over 75 greenfield GCCs in 2025, compared with 40-plus in Karnataka. “Only 30% of software engineering happens on the laptop. The real 70% starts after you commit the code,” says Jyoti

From Groundwater And Slopes To Carbon Sinks, Tools Like CatBoost

From groundwater and slopes to carbon sinks, tools like CatBoost are enabling Indian scientists to extract insights and drive sustainability. With capacity expected to more than double this decade, the industry is investing in training as graduates struggle to meet Yet despite this momentum, most respondents reported achieving satisfactory ROI on a typical AI use case within two to four years. Thi...