Moderated by Dr. Saikat Chaudhuri of UC Berkeley, the session brought together Anne Hoecker, Global Head of Technology, Media and Telecom and Board Member at Bain & Company, and Priyadarshi Ashok Das (PAD), Corporate Vice President – Telecom, Media, Entertainment & Education at HCLTech, to examine what it really takes to unlock value.
Both speakers were clear: pilots are easy. Scale is hard.
From pilots to CFO scrutiny
Hoecker drew a distinction between experimentation and enterprise impact. Organizations have spent the past two years running sandbox initiatives and isolated pilots. But that phase is ending.
As she put it, the focus now is on “really driving ROI at scale, which is what the CFO is really starting to ask for.”
Across industries, Hoecker pointed to three domains are emerging as the most mature value pools.
First is software development. Large enterprises are deploying AI coding tools across thousands of engineers, accelerating product cycles and rethinking resource intensity in R&D.
Second is services, particularly knowledge-intensive functions where AI helps employees find information faster and resolve issues more efficiently.
Third is sales, an area where structural inefficiencies are being exposed. With automation handling preparation and data synthesis, sales teams can redirect effort toward customer engagement and revenue generation.
But Hoecker was clear that AI does not automatically determine whether the benefit shows up in cost or growth. “These AI tools…are freeing up time,” she explained. What organizations do with that time becomes the strategic decision.
Some may choose cost reduction. Others may choose acceleration.
Ultimately, she framed the ambition: “I want my revenue line to grow faster than my increases in cost. And AI will let you do that.”
Why ROI fails to materialize
If the opportunity is so clear, why are enterprises struggling to realize it?
For PAD, the issue is structural.
He pointed to fragmentation inside organizations, including disconnected foundation layers, siloed platforms and unintegrated systems, which prevent AI from scaling across workflows. True transformation requires coordination across multiple players, because “It's never a one player does it all.”
Hyperscalers, ISVs, ecosystem partners and system integrators must operate within a coherent orchestration layer. Without it, enterprises end up with pockets of automation but no enterprise-wide ROI.
More fundamentally, PAD argued that many organizations have not maintained architectural discipline. When companies compromise on foundational architecture, particularly around data, they accumulate debt that limits long-term scalability.
That debt is not only technical. It is operational.
Projects are often completed without being tied back to measurable business impact. Dashboards track activity, such as tickets closed and hours logged, but fail to connect delivery to customer outcomes or financial performance.
PAD described a shift underway toward more mature engagement models, where “the business outcome is linked to the suppliers or the SI companies' parameters.” In these models, technology partners share accountability for impact, not just execution.
The rise of outcome-based partnerships
The implications extend beyond enterprise IT.
Both speakers acknowledged that AI is reshaping pricing models across the services and consulting industries. As work becomes a combination of people, agents and automation, billing purely for human effort becomes outdated.
Instead, organizations are moving toward outcome-based or at-risk models, aligning compensation with measurable gains in efficiency, revenue or performance.
This alignment, PAD argued, requires early involvement. Strategic partners must be embedded at the point where architecture and outcome design are defined. His call to action was simple: “give us a seat at the table.”
When partners are brought in late, after fragmentation has set in, ROI becomes harder to engineer.
Focus beats fragmentation
Both leaders agreed that governance and prioritization are critical.
Hoecker observed that bottom-up experimentation often leads to a proliferation of small pilots with limited enterprise impact. To scale, organizations need clarity at the top. “You need to have some top-down direction from the CEO or CEO minus one, saying we're going to focus on [specific areas],” she said.
Without that CEO-level focus, enterprises struggle to convert time saved into measurable value.
The dashboard also matters. Tracking usage alone is insufficient. Leaders must monitor metrics that tie directly to business impact, such as customer visits, resolution time and product cycle compression, and review them consistently at executive level.
That discipline transforms AI from a toolset into a strategic lever.
Beyond cost reduction
As Dr. Chaudhuri summarized in closing, ROI requires prioritization, process and shared accountability. Partners can help clients navigate the sea of possibilities, but success ultimately depends on integration and measurement.
The first wave of AI has been about efficiency. The next wave is about divergence and ensuring revenue growth outpaces cost expansion.
That shift demands architectural rigor, governance discipline and outcome-based collaboration across ecosystems.
AI may be the catalyst. But scalable ROI will belong to organizations that design deliberately, integrate systematically and align incentives around results.
