Leadership Spotlight
The Real AI Differentiators with Gururaj Bhat
Most companies are chasing AI. The leaders will be the ones that organize their data, apply AI to it intelligently, and translate that into measurable business outcomes. AI alone isn't the advantage. The advantage is knowing where to apply it, how to apply it, and how to prove its value.
A Conversation with Gururaj Bhat, Executive Vice President, Data & AI
As enterprises race to deploy AI, many are finding that access to technology is no longer the challenge. The real challenge is turning AI into business value. According to Gururaj, the answer lies not in AI alone, but in the combination of domain expertise, trusted data, and a relentless focus on outcomes. In this conversation, he shares why organizations struggle to realize value from AI, what leaders should stop measuring, and how Evalueserve is helping clients move from experimentation to transformation.
Everybody is talking about AI. What do you think most organizations are still getting wrong?
I think many organizations start in the wrong place. They start with AI because that’s where the excitement is. The reality is that AI is only as good as the data underneath it.
I’ve seen companies under pressure from leadership to roll out AI initiatives before they’ve done the hard work of organizing their data. But if the foundation isn’t there, the outcomes won’t be there either.
For me, AI has never been the starting point. The starting point is understanding the business problem, understanding the data, and then determining how AI can create value. That’s why I’ve always thought about this as the combination of domain expertise, data, and AI, not AI in isolation. AI is the accelerator, but data and business understanding are what determine whether you end up somewhere useful.
You've spoken before about "AI value realization." What does that actually mean in practice?
It’s actually much simpler than people make it sound.
Every AI initiative should answer a basic question: What business outcome are we trying to achieve?
Are we increasing revenue? Reducing costs? Helping people work faster? Improving risk management or compliance?
If you can’t clearly answer that question, you’re probably not ready to start.
What happens in many organizations is that teams deploy AI, people use it, and then six months later somebody asks, “What value did we create?” and nobody has a good answer because there was never a baseline or a measurement framework.
Value realization means defining success before you start and tracking whether you’re actually achieving it. Otherwise, you’re measuring activity instead of outcomes.
What's the biggest mistake companies make when selecting AI use cases?
They often choose the hardest problems first.
There’s a tendency to believe that if AI can solve the most complex exception case in a workflow, then it must be valuable. But those are usually the applications where the technology is least mature and the risk of failure is highest.
I think organizations need to be more pragmatic.
Look for use cases where the objective is clear, where the data exists, and where the value can be measured. Build credibility and momentum there first.
The goal isn’t to prove that AI is impressive. The goal is to prove that AI creates value. Those are two very different things.
Many companies still measure AI programs using technical metrics. What should leaders be measuring instead?
I would actually start by asking what they should stop measuring.
People talk about token consumption, model usage, prompts generated, things like that. Those metrics might tell you something about adoption, but they don’t tell you whether the business is performing any better.
Nobody wins because they used more tokens.
The metrics that matter are business metrics. Revenue uplift. Cost efficiency. Productivity gains. Faster cycle times. Better risk outcomes. Stronger compliance.
If AI isn’t moving one of those measures in the right direction, then it really isn’t creating value, regardless of how sophisticated the technology looks.
Despite widespread access to AI tools, adoption remains difficult. Why?
Because tool training isn’t the same as workflow transformation.
Many organizations provide employees with access to a model, offer some basic training, and assume adoption will happen naturally. But employees don’t work inside AI tools. They work inside business processes.
If I’m in lending, show me how AI improves loan origination. If I’m in manufacturing, show me where AI fits into the decisions I make every day.
People need to understand how AI connects to their work. They need to trust the outputs. They need confidence in the underlying data. Once those elements are present, adoption becomes much easier.
The conversation should be less about learning the tool and more about improving the process.
You've been closely involved in Evalueserve's thinking around Agentic Squads. Why does that model resonate with you?
Because transformation is never purely a technology problem.
You need people who understand the business process. You need engineers who can implement AI effectively. And you need enablement and change management expertise so people actually use the solution.
Those capabilities usually exist in organizations, but they often operate separately.
What I like about the Agentic Squad concept is that it brings those disciplines together around a business workflow. The objective isn’t to deploy technology for its own sake. The objective is to solve a problem and create measurable value. That’s a much better way to think about AI transformation.
Looking ahead, what will separate the winners from everyone else in the age of AI?
I don’t think the winners will be defined by access to AI. Eventually, everyone will have access to comparable technology.
The differentiator will be their data and their ability to operationalize AI against that data faster than competitors.
Every organization has accumulated knowledge through years of doing business. The question is whether that knowledge is organized, accessible, and usable.
The companies that can harness their unique data assets and apply AI across workflows to create outcomes will have a significant advantage.
That’s where differentiation comes from. Not from the model itself, but from what you do with it.
As AI becomes more accessible, Guru believes competitive advantage will not come from the technology itself. Because in the end, AI is not the differentiator. The differentiator is the ability to turn unique knowledge, trusted data, and human expertise into real business value.
Latest Posts
Private Equity Monthly Newsletter – Sept 2026
Co-investments, retailization and tokenization are reshaping private markets as investors seek access, liquidity and efficiency.
Auto Portability and the Rise of Seamless Retirement Transfers
Explore how auto portability reduces retirement savings leakage, simplifies 401(k) transfers, and helps workers preserve savings when changing jobs.