Leadership Spotlight
The Truth Behind AI ROI with Basab Bhattacharya
Our generation has witnessed three transformational shifts: computers, the internet, and now AI. Change is inevitable, and if history has taught us anything, it’s that resistance is futile, We shouldn't resist this change. Instead, we must learn how humans and intelligent systems can work together to create more value than either could alone.
A Conversation with Basab Bhattacharya, Vice President, Data & AI Solutions
Enterprise AI has progressed rapidly from experimentation to deployment. Yet despite unprecedented access to powerful models, cloud infrastructure, and automation technologies, many organizations continue to struggle to realize meaningful business value.
According to Basab Bhattacharya, the challenge is no longer technology. The challenge is translating AI into tangible and measurable outcomes. That requires something many organizations still underestimate high-quality data, deep domain expertise, and people who understand how business decisions are made.
Enterprise AI is more accessible than ever. What do you think organizations are still getting wrong?
Many organizations are still approaching AI primarily as a technology initiative. They invest significant time evaluating models, platforms, and tools, but they spend far less time understanding whether the underlying data and business context are ready to support the expected outcomes.
The reality is that technology is becoming a commodity, so access stopped being a differentiation; now data is.
And not simply the volume of data, but its relevance, quality, context, and accessibility.
Data issues are silently undermining the ability to scale AI in some companies. They might build a successful pilot, but when they attempt to scale it in production, they discover that data is fragmented, inconsistent, poorly governed, or disconnected from the decisions they are trying to improve. Which leads to what we call the graveyard of POCs, because data was an afterthought.
Most organizations think they have an AI problem. More often, they have a data problem.
You've spoken about domain expertise as a competitive advantage. Why is it becoming more important in the age of AI?
Because AI can scale knowledge, but it cannot replace expertise.
One of the things that makes Evalueserve unique is the depth of expertise within our organization. We have specialists, researchers, data scientists, industry practitioners, and advanced subject matter experts who understand how businesses operate, how decisions are made, and where value is actually created.
Technology alone cannot provide that context.
In fact, as AI becomes more capable, domain expertise becomes more valuable. Organizations still need people who can ask the right questions, challenge assumptions, understand risk, interpret results, and make decisions in complex environments.
As we’ve often said, domain expertise is one of the most valuable assets these days. When you combine deep industry knowledge, domain and functional expertise with strong engineering and trusted client relationships, you create something that cannot be easily replicated.
There is a lot of discussion about agentic AI and autonomous systems. What role do humans play in that future?
Humans remain at the center. I don’t see the future as humans versus AI. I see it as humans working alongside increasingly capable systems, often ‘in the lead’, sometimes ‘in the loop’, depending on the value and the risks associated.
AI is exceptionally good at processing information, identifying patterns, and executing tasks at scale. Human experts bring judgment, context, creativity, and accountability.
That’s why I believe the future of work is fundamentally collaborative, combining the two effectively. We’re moving toward environments where AI systems assist experts, accelerate decisions, and automate routine tasks, allowing people to focus on higher-value activities.
The replacement-amplification debate reflects two fundamentally different views of AI: one focused on substituting human work, the other on expanding what people can achieve. By focusing on amplification, AI and human expertise work together, and organizations can achieve outcomes that neither could accomplish independently.
Yet many AI initiatives still struggle to demonstrate business value. Why is that?
Because value realization is rarely a technology challenge.
Most organizations can prove that a model works. Many can successfully deploy a solution into production. What remains difficult is connecting those efforts to business outcomes.
Too often, initiatives are selected because they’re technically feasible rather than strategically important. Teams focus on deployment metrics instead of operational improvements. Success is measured by adoption rather than impact.
The issue is that many initiatives begin with the technology instead of the business problem.
Once you understand where value exists, how decisions are made, what processes need improvement, and which metrics matter to the business, only then you can determine how AI can contribute.
That’s where domain expertise, business understanding, and strong data foundations become critical.
Yes, many organizations still celebrate AI deployments as milestones. You argue that success should be measured differently. What should executives be measuring instead?
Today, most executives don’t need another demonstration that a model can summarize documents, generate content, or automate a task. The question they’re asking is much simpler and much harder: Where is the business value?
Is it reducing costs?
Increasing revenue and sales?
Improving productivity?
Accelerating decisions?
Reducing risk?
Ultimately, AI needs to be measured in the language the business understands. That’s not prompts, models, tokens, or deployments.
It’s dollars.
If an initiative cannot be connected to financial outcomes, it becomes very difficult to prioritize, scale, or sustain. That’s why I believe organizations need an AI value realization framework that establishes baseline metrics before deployment and measures business impact afterward.
We don’t need to prove the technology works. That debate is largely over. We need to demonstrate that AI creates measurable economic value, and how.
You spoke about how our generation has experienced several major technology shifts. How do you view the current AI moment?
I think we’re incredibly fortunate.
Our generation witnessed the widespread adoption of personal computers, the rise of the internet, and now the emergence of artificial intelligence. Each of these technologies fundamentally changed how people work, communicate, and create value.
- Computers transformed individual productivity.
- The internet transformed connectivity and access to information.
- AI is transforming how knowledge works itself gets done.
That’s why I don’t see AI as something organizations and the general populace should resist. Every significant technological shift creates uncertainty, but it also creates opportunities.
Embrace change, learn quickly, and help your people adapt, is the recipe for successful leaders.
The same will be true with AI. If people can handle it correctly, there is considerable room to expand what’s possible.
What advice would you give business leaders preparing for the next phase of AI?
Start with data.
Before thinking about models, make sure you understand your data landscape, your governance approach, and the quality of the information your organization depends on. And if your data infrastructure is ready for a large-scale AI transformation,
Second, invest in expertise.
Technology is becoming more accessible every day. Deep domain knowledge remains scarce.
Third, think about AI as a capability that enhances people rather than replaces them.
The organizations creating the most value are combining intelligent systems with talented people who understand customers, industries, processes, and decisions.
Organizations that combine trusted data, deep domain and industry expertise, and intelligent systems will be best positioned to create lasting value.
Having witnessed the rise of computers, the internet, and now AI, Basab sees a familiar pattern. Every transformative technology changes how we work, create, and compete. The real promise of AI is helping people make better decisions, move faster, and unlock value hidden within their data. The future is not humans versus AI.
Latest Posts
From Component to Connected System: Shifting Flow Control from Single SKUs to Intelligent Modular Skids
Flow control is moving from single components to connected systems. Here’s how intelligent modular skids, zero-touch commissioning, and predictive service are changing where manufacturers create value.
The Grid Behind the Growth
Explore how rising power demand is driving U.S. power grid infrastructure investment and creating opportunities across equipment, construction, and utilities.