GCCs Hit 92% AI Adoption but 72% Lack ROI Frameworks; Five-Lens Model Targets Pilot Purgatory

An AI ROI framework is a structured method for evaluating artificial intelligence investments across maturity, baselines, adoption depth, total cost of ownership, and value delivered, as defined in a new whitepaper co-developed by Zinnov and ProHance for Global Capability Centers.

India's Global Capability Centers have moved past the question of whether to adopt artificial intelligence and into a harder question: how to prove the technology pays for itself. A new whitepaper from Zinnov and ProHance, published through the Nasscom community, reports that 92% of GCCs in India are piloting or scaling AI use cases, yet 72% of leaders say they lack a structured ROI framework to measure the results. The gap between deployment and measurement is now the central constraint on scaling AI value across enterprise operations.

The whitepaper, titled

Global Capability Centers are offshore or nearshore subsidiaries that large multinational enterprises use to centralize technology, engineering, analytics, and business-process capabilities. India hosts more than 1,700 such centers, and they have become the primary proving ground for enterprise AI initiatives because they concentrate technical talent and data assets in one operating model. The Zinnov-ProHance research, based on inputs from more than 160 survey responses from leaders and employees and more than 50 hours of leader interviews, argues that GCC leaders have no shortage of use cases but remain weak at connecting those use cases to financial outcomes.

Karthik Padmanabhan, managing partner at Zinnov, framed the problem in terms of execution rather than ambition. “AI adoption in GCCs is no longer the barrier – 92% are already piloting or scaling use cases. The real hurdle is ROI,” he said. Saurabh Sharma, chief operating officer at ProHance, warned that unstructured experimentation could stall progress. “GCC leaders are not short of ambition when it comes to AI, but without a credible adoption and ROI framework, that ambition risks getting trapped in pilot purgatory,” Sharma said. The phrase captures the pattern the study documents: many initiatives remain in proof-of-concept mode indefinitely because teams cannot demonstrate business value at scale.

The study identifies four readiness pillars that determine whether a GCC can move from pilots to production value: Data & Infrastructure, Talent, Governance & Change, and Adoption & Usage Depth. Each pillar represents a distinct failure point. Fragmented data makes model outputs unreliable. Talent shortages slow the transition from experimentation to product. Weak governance creates compliance risk and erodes stakeholder trust. Shallow adoption means even technically successful models never change how work is done.

The whitepaper's central contribution is a five-lens AI ROI framework designed to replace ad hoc metrics with a repeatable evaluation process. The five lenses are maturity assessment, baselines, adoption depth, total cost of ownership, and value delivered. Together, they move the conversation from generic AI enthusiasm to specific, comparable, and auditable outcomes.

Maturity assessment asks where a GCC sits on the adoption curve and what capabilities must exist before value can be captured. Baselines establish the pre-AI performance level so improvements can be measured against a known starting point. Adoption depth examines not just whether a tool was deployed but how deeply and consistently employees use it in daily workflows. Total cost of ownership accounts for infrastructure, model licensing, data engineering, change management, and ongoing maintenance rather than only initial build costs. Value delivered ties outputs to business metrics such as productivity, cycle time, quality, and cost savings.

The framework also draws on a dual-lens methodology that compares leader surveys with employee pulse checks. The study finds that leaders and employees often perceive AI usage and impact differently. That perception gap matters because leadership may declare a deployment successful while frontline teams report that the tool is ignored, bypassed, or actively resisted. A credible ROI assessment must therefore include both management intent and operational reality.

The research gives concrete evidence that structured adoption produces measurable returns. Case studies in the whitepaper highlight real GCC examples achieving 20% to 30% productivity gains and up to $18 million in annual efficiency savings through structured adoption and visibility. These numbers, the authors argue, are not the result of exotic model capabilities but of disciplined measurement, clear baselines, and deliberate change management.

The $18 million savings figure is especially significant for GCC operating models. Centers that function as cost-optimization engines can use such outcomes to reposition themselves as value creators within the parent enterprise. When a GCC can show concrete annual savings attributable to AI, it gains credibility in budget conversations and expands its mandate beyond service delivery into strategic technology decisions.

The 66% barrier figure adds urgency. Mint's coverage of the study reports that 66% of leaders cited fragmented data, poor integration, and compliance risks as barriers to scaling AI. Data fragmentation is not merely a technical inconvenience; it directly undermines the baselines and measurement the ROI framework depends on. Without integrated data, even a well-designed evaluation model cannot produce trustworthy results.

Compliance risk is amplified by the regulatory environment. GCCs in India increasingly handle regulated workloads for financial services, healthcare, and other sectors, and AI models that touch personal data introduce audit and explainability obligations. The whitepaper's governance pillar treats compliance not as a constraint but as a design input for any AI deployment that expects to survive enterprise review.

The framework is structured for practical application rather than academic completeness. Each lens maps to a set of questions that a GCC chief operating officer, chief technology officer, or AI program leader can ask before a project moves from pilot to scale. The result is a checklist that can be used consistently across different business functions and geographies.

A table summarizing the five lenses illustrates how the framework converts abstract ROI goals into operational checkpoints:

Lens	Focus	Key evaluation question
Maturity assessment	Adoption stage and capability readiness	What capabilities must exist before value can be captured?
Baselines	Pre-AI performance measurement	What is the starting point for comparison?
Adoption depth	Usage frequency and workflow integration	Are employees actually using the tool in daily work?
Total cost of ownership	Full lifecycle cost of AI systems	What does the solution cost including maintenance and change management?
Value delivered	Business and financial outcomes	What productivity, quality, or cost savings did the deployment produce?

The distinction between adoption depth and simple deployment is one of the whitepaper's most useful insights. Many organizations track whether a model is technically live but never measure whether employees integrate it into core workflows. The dual-lens employee pulse component addresses this by capturing usage patterns, friction points, and perceived value from the people who interact with AI systems daily.

For GCC leaders, the immediate implication is that AI ROI must become a governance discipline, not a retrospective exercise. The study recommends establishing baselines before deployment, defining success metrics at the use-case level, and creating visibility mechanisms that allow leadership to track adoption and value in near-real time. ProHance, which provides workforce analytics and productivity platforms, positions its technology as an enabling layer for exactly this kind of visibility.

The 72% measurement gap also has financial consequences. Enterprises that cannot quantify AI returns are less likely to secure continued investment, and pilots that cannot demonstrate value are more likely to be defunded in budget cycles. In an environment where chief financial officers are scrutinizing technology spending, the absence of a structured ROI framework becomes a direct competitive disadvantage.

The market context matters. GCCs are no longer just cost centers; they are increasingly chartered with building enterprise-wide digital capabilities. The whitepaper's findings suggest that AI investment will accelerate where measurement frameworks exist and stall where they do not. That divergence could create a measurable difference in enterprise performance over the next two to three years.

The study's methodology adds confidence to its conclusions. With more than 160 survey responses from leaders and employees and more than 50 hours of leader interviews, the research captures both breadth and depth. The dual-lens design also reduces the risk of leadership optimism skewing the results, because employee responses provide an independent check on deployment claims.

The practical path from pilot to proven ROI follows a sequence that GCCs can execute within existing operating structures. The whitepaper's framework implies an ordered approach rather than a simultaneous transformation:

1. Assess current AI maturity across the four readiness pillars: data and infrastructure, talent, governance and change, and adoption depth.
2. Establish performance baselines for each target use case before deployment so improvements can be attributed credibly.
3. Define adoption-depth metrics that track actual employee usage, not just technical deployment status.
4. Calculate total cost of ownership including infrastructure, model licensing, data engineering, and change management.
5. Measure value delivered against baselines and report outcomes in financial terms such as productivity gains or annual savings.

The order matters. Starting with maturity assessment prevents teams from investing in advanced tooling before foundational data capabilities exist. Establishing baselines before deployment prevents attribution disputes later. Defining adoption metrics early forces organizations to confront the human dimension of AI change rather than treating it as an afterthought.

The whitepaper's case studies are intended to serve as templates. The 20% to 30% productivity gains and up to $18 million in annual efficiency savings are presented as evidence that structured adoption and visibility produce results, and the authors encourage GCCs to build similar visibility into their own AI programs rather than relying on anecdotal success stories.

There are limitations to what the public summary reveals. The whitepaper's full methodology, including how productivity gains were measured and how the $18 million figure was calculated, sits in the complete document rather than the community summary. Enterprises adopting the framework should expect to adapt the five lenses to their own financial reporting standards and operational metrics.

The implications extend beyond India. Multinational enterprises treat GCCs as centers of excellence for AI because the talent pool and data assets are concentrated there, and the measurement frameworks developed in India can be exported to other global delivery locations. A GCC that demonstrates credible AI ROI becomes a reference point for the entire enterprise technology organization.

The study also speaks to the broader enterprise AI market. The 92% adoption figure confirms that AI is embedded in enterprise operations, while the 72% measurement gap explains why so many AI initiatives fail to scale. The missing ingredient is not model capability but management discipline. The five-lens framework addresses that gap directly.

For technology buyers and vendors, the findings suggest that demand for ROI measurement tools, workforce analytics platforms, and governance frameworks will grow. ProHance's participation in the whitepaper signals that workforce analytics vendors see measurement as a growth market. The enterprise AI stack is expanding beyond models and infrastructure into the measurement layer that makes AI investment defensible.

The whitepaper's governance and change pillar deserves particular attention from chief information security officers and chief compliance officers. The 66% of leaders who cited compliance risks as a barrier will need frameworks that build audit trails, data lineage, and model documentation into the AI lifecycle. A five-lens approach can accommodate those requirements because value measurement depends on traceable inputs and outputs.

GCC leaders who have not yet adopted a structured ROI framework face a narrowing window. As more centers publish quantified outcomes, parent enterprises will begin comparing GCC performance on AI value creation. The centers that can show 20% to 30% productivity gains and annual savings in the millions will attract expanded mandates; those that cannot will remain in the cost-center lane.

The next phase of GCC AI development will likely focus on standardizing ROI metrics across use cases. The whitepaper provides a starting framework, but the market will need industry-specific benchmarks, shared definitions of productivity, and common methods for calculating total cost of ownership. The organizations that contribute to building those standards will shape how enterprise AI value is measured for years.

ProHance's Sharma has stated the risk directly: without a credible adoption and ROI framework, ambition gets trapped in pilot purgatory. The study offers a way out, but it requires leaders to treat measurement as a first-class activity rather than a post-deployment report. That shift in mindset is the difference between AI as a portfolio of experiments and AI as a driver of scalable enterprise value.

The evidence from the whitepaper is that the path to proven ROI is neither mysterious nor dependent on frontier model capabilities. It depends on baselines, adoption depth, total cost of ownership, and value delivered, all assembled into a repeatable five-lens review. GCCs that operationalize that framework will be the ones that convert AI adoption into durable financial performance.

Enterprises evaluating their own AI measurement practices can use the study's findings as a diagnostic. A center that cannot answer basic questions about baselines, adoption depth, and total cost of ownership should treat that gap as a risk before expanding its AI portfolio. The 72% of leaders who lack a structured ROI framework are not outliers; they are the majority, and the market is now offering them a credible template for change.