As enterprise leaders work to move from AI adoption to full AI ROI, these questions are coming up most often in their coaching conversations.
Sara Braun
Editorial Strategist, HR
Workday
AI is everywhere—but value is not.
For years, adoption metrics were enough: How many people were using AI, how many pilots were live, how many departments were leveraging tools. Today, leaders are expected to show the return on AI investment in a more impactful way, proving whether its implementation is actually contributing to business outcomes.
Organizations with AI-powered workflows are twice as likely to report meaningful time savings, according to Workday’s research. However, only about a quarter (27%) of organizations have embedded AI directly into core business processes.
That challenge is showing up in coaching conversations and executive forums, including Workday Elevate in San Francisco, where leaders described today’s workforce as “human APIs” whose days are still largely spent moving information from pile A to pile B.
Chief Learning Officer Chris Ernst put it best: AI’s real promise is to flip this script so that people are no longer buried in manual work, but free to exercise judgment, creativity, and leadership with AI as a fast and intelligent tool to augment their workflows.
Monica Bua, an executive coach and Censia Executive Intelligence president, noted that leaders are now firmly positioned at the helm of this transformation, shifting from commanders to “systems architects” who orchestrate end-to-end workflows, technology, and teams, and adapt them quickly and continuously as conditions change.
So how can leaders actually get from AI adoption to a truly AI‑native operation? In recent leadership and career coaching sessions, this has emerged as one of the biggest questions.
Report
Leaders are moving from commanders to systems architects who orchestrate workflows, technology, and teams.
Employees are motivated and care deeply about their work, but they're often stuck in outdated role designs. Workday research shows that while eight in ten employees say they spend too much time on coordination, 97% of those who use AI still rate their day-to-day experience as positive.
This gap points to a role problem—not a motivation problem—and leaders are increasingly asking how they can move from simply adding AI on to existing jobs to redesigning the work around new AI capabilities.
It starts with clarifying AI versus human responsibilities. Teams need a shared answer to "Who does what?" across each workflow—where AI will automate or assist, and where human judgment is still a non-negotiable.
From there, leaders can map new AI-augmented workflows. Instead of bolting tools onto old processes, they can start with the desired outcome and design workflows backward, determining the best combination of people and AI to achieve it.
2. What Skills Does Our Workforce Need?
Employees are already using AI to remove friction (43%) and reduce manual work (46%), yet uneven access to skills and training still limit value. Employees and leaders both want to know which capabilities will matter most as AI becomes central to the operating model.
Some capabilities are becoming foundational, like AI literacy, basic data skills, promptcraft, and the ability to work with embedded AI tools in daily systems. These ensure that everyone can participate successfully in AI-supported workflows.
At the same time, leaders shared with Workday that they believe distinctly human skills are becoming more valuable as AI becomes more prevalent. Problem-solving, communication, systems thinking, and leadership remain critical in AI-rich environments.
As Bua pointed out, "tacit skills" such as judgment and creativity are gaining importance exactly when evaluating them at scale is becoming more difficult.
Leaders who communicate clearly which skills they'll reward—and back that with investments in learning, coaching, and stretch opportunities—create a powerful signal to employees that AI is a path to growth, not obsolescence.
Distinctly human skills are becoming more valuable as AI becomes more prevalent.
Employees still feel connected to purpose—88% say they see how their work contributes to broader outcomes—but they're too often operating in systems designed for a pre-AI world.
However, career structures are slowly beginning to shift. Organizations are becoming more fluid, with leaders expecting fewer management layers and smaller, cross-functional teams with broader remits instead of narrow functional silos.
New "AI-plus-human" roles are also emerging. From AI product owners to AI automation architects, roles that blend domain expertise with AI fluency are becoming pivotal.
Career paths are becoming more lattice than ladder as well. Internal talent marketplaces and AI gig opportunities are enabling employees to build skills across functions. At Workday, for example, more than 6,000 internal gigs have driven a 43% increase in internal mobility and a 13% lift in long-term career sentiment for participants.
Leaders who explicitly map and communicate future career paths in an AI era—not just for tech roles, but for every function—can counter uncertainty and help retain top talent.
Again, only about a quarter of organizations have embedded AI directly into core workflows—yet those that do are nearly twice as likely to report meaningful time savings.
Chris Ernst offered a cautionary perspective. Three years ago, the average Fortune 500 company had zero AI agents. By 2030, analysts estimate they'll have more than 150,000 agents running. Simply "bolting on an ever-growing array" of agents, he warned, doesn't work; it risks turning employees into coordinators of bots rather than strategic contributors.
Organizations seeing real impact take a different path. They start with the highest-friction workflows and redesign them so that AI sits inside the flow of work, not on top of it.
Finally, meaningful scale requires broader measures of success. Leaders track not only time saved but also improvements in decision quality, customer experience, and employee engagement. As Bua emphasized, the goal is elevation, not just efficiency—moving people from administrative tasks into high-judgment, high-impact work.
Organizations with AI embedded in their workflows are twice as likely to report meaningful time savings.
Human oversight is most critical where the stakes are highest: decisions that materially affect people, money, risk, and brand. Employees understand this intuitively.
When workers trust the underlying systems and data, 87% say AI increases their confidence in decisions, and 89% of operations and IT professionals say AI is more valuable when it runs on a trusted operational core.
At the same time, barriers such as poor data quality, unclear guidance on when to use AI, and outputs that conflict with existing policies all limit value. That’s why leaders need a clear framework for which decisions can be automated and which must remain human‑led.
Broadly, decisions fall into three categories:
In practice, this means high‑risk functions like HR, finance, legal, customer service, and marketing need explicit guardrails. For example, AI may be able to draft communications but not distribute them, or screen job candidates but not single‑handedly reject them. In finance, AI might recommend expense anomalies or forecast scenarios, but humans should be the ones to determine policy exceptions or large capital allocations.
Oversight is also essential across the lifecycle of AI initiatives, not just at the moment of decision. Privacy, security, and compliance experts should be involved from the start rather than brought in as a final checkpoint, and be part of the teams that monitor and assess performance over time.
That keeps governance at the center of ongoing AI optimization—and makes it clear where AI can safely take the lead versus where humans must remain firmly in charge.
Many employees feel overwhelmed by tool sprawl. Even among organizations scaling AI, 82% of employees still say navigating processes and systems is a source of stress, and nearly half have been forced to work around core systems altogether.
The right AI tools for your business are the ones that reduce this friction instead of adding to it. They do three things especially well:
As Bua noted, boards shouldn’t be asking only, “How do we get more efficiency out of AI?” They should also ask, “If information is no longer scarce, how should we be thinking about our company?” That reframing points leaders toward systems‑level choices—choosing AI that reinforces a unified, trusted operating core—rather than accumulating one‑off tools that look impressive but don’t change how work actually gets done.
As AI touches more sensitive processes across HR and finance, leaders are increasingly concerned about data security, regulatory exposure, and reputational risk. When AI runs on a trusted operational core, its value increases precisely because strong governance protects the integrity of decisions—not just their speed.
Achieving privacy and compliance at scale starts with where governance sits. Instead of treating legal, security, and compliance as a final checkpoint, mature organizations bring these partners into the earliest stages of AI design: defining use cases, selecting data sources, deciding where AI can act autonomously versus only recommend.
That early involvement allows teams to encode policies, approvals, and audit requirements into the systems AI uses, so compliant behavior is the default path rather than an afterthought. From there, the work is about making guardrails actually usable.
Teams beyond legal and security need clear, practical guidance on what data can be used for which purposes, what transparency is expected when AI informs a decision, and how to escalate concerns when something doesn’t look right.
That guidance should show up inside tools and workflows themselves—through prompts, role‑based access, and approval flows—so people don’t have to remember a separate rulebook every time they use AI.
Ultimately, privacy and compliance become scalable when they are treated as design inputs that shape how AI interacts with core systems, data, and people. The question shifts from “How do we keep AI from breaking the rules?” to “How do we design AI so following the rules is built in?”
When AI runs on a trusted operational core, its value increases because strong governance protects the integrity of decisions.
The leaders at Elevate left with a powerful challenge from Bua: “Stop workforce planning and start capability planning.”
In an AI‑native world, that means asking:
As Ernst reflected, leaders in this moment are “more mentally and intellectually challenged than ever,” but this challenge also represents an unprecedented opportunity.
HR, finance, IT, and business leaders have a chance to move beyond tracking AI adoption to architecting AI advantage, where the technology is woven into core workflows, employees are growing into new, judgment‑rich roles, and trust is built into every decision.
That’s the path from AI adoption to true AI maturity—and the leaders who start answering these questions now will demonstrate lasting business impact.
Report
A gap between effort and impact is the real ROI problem. Discover the solution.