Indian companies are racing to adopt artificial intelligence, anticipating a new era of productivity. Yet, a landmark study from Cognizant reveals a surprising paradox: while employees feel the benefits, most firms are struggling to see a financial return.
The Productivity Paradox
A new study by Cognizant highlights a significant disconnect in India's booming AI landscape. On one hand, employees are embracing the technology. The research, which surveyed 700 enterprise employees and 75 senior executives, found that 97% of employees report productivity improvements from using AI, with an average gain of 15%. Indian workers are also proactive in their preparation, with 74% having completed AI training, a figure significantly higher than the global average of 56%. However, this personal productivity boost is not yet translating into measurable business success. The same study reveals that a staggering 91% of business leaders report no impact on revenue from their AI initiatives, and 84% have not seen a reduction in costs. This has led to significant uncertainty, with 37% of leaders admitting they have paused or even stopped AI projects due to an unclear return on investment (ROI).
Why the ROI Remains Elusive
The core of the problem is not that AI isn't working, but that its impact is incredibly difficult to measure with traditional methods. One of the primary hurdles is the challenge of isolating AI's contribution. When AI is integrated into complex workflows, its benefits—like improved decision-making or faster problem-solving—get blended with other business activities, making it hard to assign a specific monetary value. Furthermore, many firms lack a clear strategy and the baseline metrics needed to track progress. According to a separate report by Avanade, 74% of Indian organisations have not yet formally calculated AI ROI in a way that is reviewed by finance teams or senior leadership. This is compounded by a lack of ownership; when no single department is responsible for tracking AI outcomes, pilots rarely scale because no one can definitively prove they were successful.
The Data and Skills Gap
Even with a will to measure, many companies lack the means. Effective AI relies on high-quality, well-structured data, and many Indian firms are still grappling with this foundational layer. A Dun & Bradstreet survey found that while nearly 70% of Indian businesses plan to increase AI spending, only 4% consider their enterprise data fully ready to support AI at scale. Integrating AI with legacy systems is another major challenge, cited by 48% of respondents in the Avanade study as a significant barrier. Beyond technology, there's a human element. While employees are training themselves, there is a deeper shortage of specialised talent needed to build and manage the complex frameworks for monitoring AI performance. The speed of AI deployment in India has outpaced the development of in-house expertise required to fine-tune models and extract long-term value.
What High-Performing Firms Do Differently
Despite the challenges, some companies are beginning to see returns. A recent Kyndryl report noted that 60% of Indian business leaders claim to be seeing positive returns on their AI investments. These firms often share common traits. They move beyond vague goals and define clear, measurable key performance indicators (KPIs) before a project even begins. Instead of focusing only on broad financial metrics like revenue or cost, they track more specific operational improvements, such as time saved on a specific task, reduction in error rates, or an increase in customer satisfaction scores. They also ensure that there is clear organisational alignment, with technology and business leaders agreeing on the AI strategy. These leaders understand that proving value is the critical next step after a successful pilot, and they build the business case for scaling AI initiatives with concrete evidence rather than just enthusiasm.
