Gautam Singh, Global Business Unit Head of Analytics, Data and AI, WNS Analytics – Interview Series

Gautam Singh is the Business Unit Head of WNS Analytics and the Co-founder & CEO of The Smart Cube, a WNS Company. He spent 20 years establishing and growing The Smart Cube (a research and analytics leader) before it was acquired by WNS. Prior to this, he worked for 10 years in management consulting and venture capital in Europe and the US. Gautam has held various roles, including positions at Coven Partners (London), A.T. Kearney (London), Mitsubishi Motors (India), and Cummins Engines (US). He holds an MBA from the University of Michigan, Ann Arbor, USA and an undergraduate degree in Mechanical Engineering from IIT Bombay, India.

WNS Analytics helps companies turn their data into strategic value through “decision intelligence”—combining robust data infrastructure, AI/GenAI/agentic-AI technologies, and domain-specific expertise. They offer services across industries including insurance, banking & financial services, retail, CPG, manufacturing, healthcare, energy, and logistics. Their capabilities span data engineering and governance, descriptive and predictive analytics, AI/ML tools, and visualization — all designed to enable faster, more confident decisions and continuous innovation.

You started your career in top-tier management consulting, then founded The Smart Cube and led it for over two decades before its acquisition by WNS. What motivated your transition from consulting to entrepreneurship, and later into leading a global data analytics and AI business?

I spent ten years in management consulting and saw a clear market gap: companies were sitting on mountains of data but not extracting its full value. Back in 2003, analytics was still basic – we were working with Excel spreadsheets.

The decision to leave corporate life came down to self-belief. I saw an opportunity to help organizations truly harness their data, so I founded The Smart Cube with that vision.

After 20 years of building The Smart Cube, joining WNS wasn’t an exit but an evolution. I’ve carried forward the entrepreneurial mindset, but now with far greater resources and reach. This allows us to tackle problems at a scale I could never have delivered in a smaller business. Most importantly, I recognized the power of embedding and infusing data and analytics into core business processes rather than treating them as discrete interventions. That seamless integration of domain expertise and process transformation is central to WNS’ DNA – and it’s what motivated me to be acquired and now to lead this business unit at WNS.

In your 20+ years working in analytics, how have you seen the role of data and AI in financial services evolve—from early adoption to today’s large-scale, enterprise-level integration?

In the late 90s, analytics meant looking at historical data and making statistical forecasts. The transformation has been remarkable.

The early 2000s ushered in digitization and more advanced predictive models. By 2010, real-time trading analytics had become standard. Nearly a decade ago, machine learning began driving a real shift, and more recently, Generative AI (Gen AI) has taken center stage.

Today, financial institutions treat data as a strategic asset. The question has moved from “can we use AI?” to “how do we embed AI into every decision?”

The impact is tangible: customer onboarding that once took days now finishes in hours with AI-based verification. Credit risk assessments evaluate hundreds of real-time data points beyond traditional scores. Risk calculations that required overnight batch runs are now instantaneous. And fraud detection no longer reacts after the fact – it blocks suspicious activity in real time.

How are forward-looking enterprises using AI-driven data lakes and governance frameworks to improve real-time decision-making, regulatory compliance, and transparency in financial operations?

Building monolithic data warehouses and hoping for insights no longer works. Institutions need to design intelligent data ecosystems.

Financial services face a unique challenge: they are customer-facing, handle highly sensitive data and still need to deliver personalization and real-time responsiveness. This calls for modular data lakes built on flexible frameworks.

Within this architecture, organizations create specialized data ponds for pricing analytics, risk assessment and regulatory reporting. Each pond operates independently while feeding into the larger ecosystem, delivering immediate value while preserving security boundaries.

The Zero ETL trend is especially relevant here, as it eliminates complex Extract-Transform-Load processes by enabling direct querying across systems. This allows AI to access and analyze data in real time without moving it, reducing latency and maintaining governance.

AI agents are also evolving beyond anomaly detection. They not only flag suspicious transactions but also recommend actions and execute responses within governance parameters. In compliance, AI continuously monitors transactions, generates reports and identifies issues before regulators do.

Synthetic data is often touted as a secure way to train AI models without exposing sensitive information. Can you share examples of how synthetic data is being applied effectively in fraud detection, risk analytics, and model validation?

At WNS Analytics, we leverage advanced synthetic data generation to create high-fidelity, privacy-compliant datasets that accelerate AI model training, particularly in data-scarce domains. Our synthetic datasets emulate real-world scenarios while reflecting the same statistical patterns, behaviors and correlations as actual financial data – transaction flows, fraud trends, customer behaviors – without exposing any sensitive Personally Identifiable Information (PII) or customer data.

This capability is transforming financial services across fields such as risk analytics, fraud detection, credit scoring, stress testing and compliance modeling. These synthetic datasets enable organizations to rapidly jumpstart AI solution development while ensuring both data privacy and regulatory confidence.

A particularly innovative application involves using PII-masked data to create lookalike models. This allows companies to push targeted offers to consumers, enabling personalized marketing while maintaining complete privacy.

Intelligent automation and AI agents are increasingly being embedded in business workflows. What are the most transformative use cases you’ve seen in financial services, and how do they improve operational resilience and performance?

Intelligent automation leveraging AI agents is accelerating enterprise workflows, enabling organizations to streamline operations and make faster, more informed decisions. These agents combine automation with advanced reasoning to deliver resilience, scalability and performance improvements.

At WNS Analytics, we apply the GAIN framework (our proprietary framework for Agentic AI implementation) to assess the right levels of autonomy for agentic AI. We further provide reusable, microservices-based components for hyperspecialized agents through our award-winning AI Utilities Hub.

In insurance, we have transformed multiple workflows through agentic AI. In motor claims subrogation, our Gen AI-powered third-party recovery detection solution, driven by autonomous agents, has achieved 85% accuracy, doubled recovery volumes and enhanced annual recoveries by approximately 49% – unlocking millions in opportunities that were previously overlooked.

In underwriting, our agentic AI-powered research assistant employs multiple specialized agents to break down complex queries, extract data from multiple sources and generate insights with 99% accuracy while reducing turnaround time by 85%.

For a leading bank, our Gen AI solution cut adverse media screening time 60% and reduced false positives 12-15%.

We also have a Gen AI-powered knowledge management solution – designed as a horizontal platform – to redefine how enterprises retrieve, reason through and contextualize vast unstructured data. By delivering precise, compliant and consistent insights in real time, it enhances decision-making, improves efficiency and strengthens operational resilience across industries.

These solutions augment human judgment, creating faster, more accurate systems.

For enterprises aiming to scale AI initiatives, what are the biggest barriers—technical, cultural, or strategic—and how can leaders overcome them?

The biggest barrier to scaling AI isn’t technology – it’s organizational readiness.

First, there are data silos across legacy systems. Full replacement is not always practical; instead, the focus should be on building intelligent bridges. At WNS, we have created “bridge teams” that pair legacy administrators with cloud engineers, accelerating implementation while preserving critical business rules.

Second, the skills gap. Enterprises need the right mix of domain experts, data engineers, data scientists and translators who can connect technical insights to business value.

Third, the pace of technological change. Our WNS AI Lab enables organizations to experiment with emerging technologies and build proof-of-concepts before committing to full-scale deployment.

On the cultural front, success depends on effective change management. We design frameworks that help employees view AI as additive rather than replacement-oriented. Establishing an AI council is also a smart move, providing governance, cross-functional alignment and a structured path for moving from pilots to enterprise-wide scale.

With growing scrutiny on AI ethics, bias, and transparency, how can financial institutions strike the right balance between innovation and responsible AI governance?

Innovation and responsibility are not opposing choices – responsibility must be built into innovation from the start.

Financial institutions need robust AI governance frameworks. At WNS, we implement frameworks that ensure AI is developed responsibly, ethically and securely. Our approach embeds checks for bias, fairness, custom KPIs and monitoring model drift. This builds trust, not just regulatory compliance.

Transparency is especially critical in financial services. If AI denies a loan, applicants deserve clear and understandable explanations.

Ultimately, responsible AI is a competitive advantage. Banks that demonstrate fairness, transparency and security in their AI systems earn customer trust. Those that treat governance as an afterthought risk regulatory penalties and reputational damage that is far harder to repair.

In the next 3-5 years, which emerging AI capabilities or data strategies do you believe will have the biggest impact on how financial organizations operate?

Three developments will reshape financial services over the next three to five years.

First, agentic AI will move from experimental to essential. Autonomous AI agents will execute complex workflows and orchestrate entire departments alongside human teams.

Second, continuous learning systems will become standard. AI will adapt from every interaction, enabling truly personalized financial services that evolve with each customer’s changing needs.

Third, we will see powerful technology convergence: quantum computing for advanced risk calculations, blockchain for transparent AI decision logs and edge computing for instant localized decisions. Together, these technologies will unlock entirely new forms of financial services we are only beginning to imagine.

Having navigated entrepreneurship, acquisition, and now a global leadership role, what guiding principles have helped you make decisions and lead teams through change?

Three principles guide me.

First, perseverance over perfection. When we started The Smart Cube, we did not have all the answers. We made mistakes, adapted and kept moving forward. Persistence with adaptability has been essential.

Second, build lasting value, not quick exits. A professor from business school once advised me—years after I had founded The Smart Cube—“Don’t focus on the exit. Focus on building a successful business that will last.” That long-term mindset has shaped every decision I have made.

Third, enjoy what you do. I have always believed that if I am not having fun, I will move on to something else. After 30 years, I still wake up excited, and that enthusiasm inspires teams through change.

Leading through acquisition reinforced another truth: change succeeds when you bring people along. Technical integration is straightforward; cultural integration – building a shared vision – is where real leadership matters.

For professionals who want to shape the future of AI in finance, what skills, mindsets, or experiences do you think will be most valuable?

The future belongs to those who can bridge worlds. Pure technical skills or domain expertise alone will not be enough.

First, develop systems thinking. Start with the market need – a clear use case – and work backwards. AI in finance requires seeing how everything connects: how a change in risk models impacts customer experience or how automation opens up new opportunities.

Second, cultivate disciplined practicality over idealism. Be excited about new technologies, but rigorous in evaluating them. Not every problem needs AI – sometimes, simple analyses or even spreadsheets can do the job.

Third, build translation skills. This is vital. Being able to explain complex AI concepts to board members and translate business requirements for data scientists is invaluable. The strongest AI leaders align technology with business strategy.

Finally, embrace continuous learning. Tools that were cutting edge five years ago are already outdated. Staying curious, humble and committed to learning will open doors to opportunities we cannot yet imagine at the intersection of AI and finance.

Thank you for the great interview, readers who wish to learn more should visit WNS Analytics.