Mankiran Chowhan says agentic AI success depends on unified enterprise data and context, with ROI-driven use cases, connected ecosystems, and rapid deployment.
As organisations push agentic AI into customer engagement, sales, and operational workflows, the real challenge is no longer deploying models, but ensuring AI systems understand enterprise context, work across connected data environments, and deliver measurable business outcomes instead of isolated proofs of concept.
In an exclusive interaction with CIO&Leader, Mankiran Chowhan, Managing Director, Sales & Distribution, Salesforce India discusses why most AI pilots fail due to fragmented data and weak contextual grounding. She explains how Salesforce is helping Indian enterprises build connected AI ecosystems through unified data architectures, headless enterprise models, and faster deployment frameworks, while also focusing on tangible ROI, operational flexibility, and scalable AI adoption across industries.
CIO&Leader: You mentioned that nearly 95% of AI pilots fail because they lack company-specific context. How is Salesforce India helping enterprises bridge this gap and make agentic AI deployments successful?
Mankiran Chowhan: 95 percent of AI pilots fail largely because they are not grounded in the context of enterprise data. In many cases, organisations run pilots using limited datasets or isolated use cases. However, for AI systems to work effectively, they need access to complete organisational context.
For example, context is not limited to a customer’s name alone. The entire relationship, behavioral history, interactions, and enterprise knowledge surrounding that individual become important. The same principle applies to agentic AI systems.
The context behind purchasing a car is different from buying jewelry or applying for a home loan. Every customer journey carries unique behavioral and operational context. Our role is to leverage unified data and connected systems to ensure AI systems understand that context accurately and operate meaningfully within each enterprise environment.
CIO&Leader: Enterprises are increasingly looking for measurable business returns from AI rather than experimentation alone. What separates organisations achieving tangible ROI from those still struggling with implementation bottlenecks?
Mankiran Chowhan: The difference lies in the use case itself. Organisations achieving strong ROI are directly tying AI initiatives to measurable business outcomes such as increasing revenue, improving lead conversion, reducing operational costs, or accelerating productivity.
The enterprises struggling with AI are often deploying it as experimentation or to demonstrate that they are participating in the AI trend. However, when AI initiatives are tied directly to top-line growth or bottom-line efficiency, measurable ROI becomes significantly easier to achieve.
If an AI use case helps generate more leads, improve qualification, or increase conversion rates, the business value becomes immediately visible. But if the initiative is largely qualitative without measurable business alignment, quantifying ROI becomes far more difficult.
The key differentiator is understanding the business purpose behind the AI deployment.
CIO&Leader: Salesforce has made multiple acquisitions over the past few years to strengthen its AI and data capabilities. How does this consolidation simplify enterprise adoption, particularly for Indian organisations?
Mankiran Chowhan: Salesforce has completed nearly 75 acquisitions since the company was founded, and approximately 17 to 18 of those have happened within the last three years. Much of this recent focus has been centered around data, AI, and connected enterprise systems.
The less fragmented enterprise technology becomes, the more powerful AI systems become.
A simple analogy is Lego blocks. When all the blocks come from the same ecosystem, they fit together seamlessly. You can expand horizontally, vertically, and build continuously. But if the blocks are disconnected or incompatible, integration becomes difficult. The same applies to enterprise systems and enterprise data.
When data moves across disconnected platforms, delays and inefficiencies emerge. However, when systems are deeply connected, organisations can derive significantly greater value from AI. Whether it is Informatica supporting governance and data management, or Qualified helping SDR agents improve pipeline qualification, connected systems improve the effectiveness of the overall enterprise architecture.
Indian enterprises are increasingly seeing the value of tightly connected platforms where governance, trust, AI, and operational systems function together seamlessly.
CIO&Leader: Salesforce is introducing headless architecture and open MCP-based frameworks. How are you helping traditional enterprises understand and adopt these newer architectural models?
Mankiran Chowhan: The concept is actually simpler than many enterprises initially assume. Organisations that already have existing technology investments can continue leveraging those systems while benefiting from connected architecture and unified data capabilities.
What changes is the flexibility of the interface layer. Enterprises are no longer restricted to a single UI or browser-driven environment. If organisations already have preferred interfaces or workflows, they can continue using them while still leveraging Salesforce capabilities underneath.
This flexibility makes adoption easier for Indian enterprises, including those operating on legacy infrastructure.
Trust and governance remain central to everything we build. Customer trust continues to be our number one value. Once enterprises know the architecture is grounded in governance, flexibility, and secure data management, the transition becomes significantly easier.
CIO&Leader: For organizations still operating on traditional systems, does adopting Headless 360 simplify deployment and reduce costs, or does it still require large-scale rip-and-replace modernization?
Mankiran Chowhan: Headless architecture essentially means decoupling the user interface from the underlying architecture. Organisations can continue leveraging their existing applications, workflows, and interfaces while still utilising Salesforce’s connected architecture, contextual data layer, and AI capabilities underneath.
Salesforce has always maintained an open architecture design. Some customers may prefer Anthropic, while others may choose open ecosystems or their own interfaces. The objective is not to lock customers into a single Salesforce UI, but to allow them to leverage the underlying architecture and enterprise data regardless of the interface they choose.
If an organisation already has a mobile application or an existing workflow environment, they should still be able to leverage Salesforce architecture without replacing everything entirely. Customers may already be using Salesforce for sales, service, or marketing, but over time they may decide they want a different interface or customer experience layer. We do not want the interface itself to become the limitation preventing organisations from utilising the value of their data.
For enterprises running legacy systems, the key requirement remains access to enterprise data. Organisations still need to extract value from existing systems and make that data accessible. What Salesforce is doing is simplifying how enterprises can leverage existing data through modern architecture layers instead of forcing complete replacement of legacy infrastructure.
Cost is one element in enterprise decision-making, but it is not the only factor. Organisations evaluate transformation decisions based on operational flexibility, customer experience goals, governance, scalability, and long-term business value.
As demonstrated through the customer examples showcased during the event, enterprises across industries and of varying sizes are already adopting it successfully without completely restructuring their technology environments or balance sheets.
CIO&Leader: Deployment timelines across the technology industry are shrinking rapidly. How has this changed CIO expectations around AI adoption and business outcomes?
Mankiran Chowhan: Enterprise expectations have changed dramatically. Earlier, large-scale transformation projects could take years before organisations saw measurable outcomes. Today, enterprises expect value in weeks rather than years.
Organisations are increasingly asking why innovation cycles should remain slow when technology has become capable of delivering rapid deployment and fast experimentation.
Large-scale transformation projects could take years before organizations saw measurable outcomes. Today, enterprises expect value in weeks rather than years.
Previously, a business leader would identify a problem, involve technology teams, search for vendors, build implementation plans, and then move toward deployment over extended timelines.
Now organisations can move from idea to production much faster. The advantage of rapid deployment is that success becomes easier to demonstrate. Once enterprises see measurable results from one successful use case, scaling additional deployments becomes easier internally.
For CIOs, CDIOs, and IT leaders, the ability to quickly prove business value has become significantly more important than large theoretical transformation roadmaps. The enterprise mindset today is increasingly centered around rapid experimentation, quick deployment, and immediate measurable impact.
CIO&Leader: Across the examples showcased — from BFSI and airlines to manufacturing and real estate — where are you seeing the strongest demand for autonomous agents in India?
Mankiran Chowhan: The demand is coming from virtually every sector. Indian enterprises are no longer approaching AI sequentially by waiting for one industry to mature before adopting it themselves.
The use cases have become simpler to deploy, and the technology has become significantly faster and more accessible.
What matters now is not the industry itself, but the relevance of the use case and the business outcome it creates.
The real estate use case is very different from the airlines example, yet both are generating meaningful business value through AI-driven transformation. Indian enterprises today recognise the scale of opportunity available through AI, and organisations across sectors are actively leaning into that transformation journey.