Ling-yi Tsai, a veteran in the HR technology space, has spent decades guiding organizations through the seismic shift from administrative software to strategic intelligence. Her expertise lies in harnessing HR analytics and integrated technology to transform how businesses manage their most critical asset: their people. From recruitment to talent management, she has seen firsthand how data is elevating the HR function from a service center to the very nervous system of the enterprise. This conversation explores the practical steps for navigating this transformation, delving into the shift from static records to dynamic workforce signals, the difference between simple automation and sophisticated AI-orchestration, and how HR leaders can become “workforce economists.” We’ll also touch on the dangers of fragmented tools and the rise of a unified “HRtech Operating System” that turns people data into a core driver of business value.
Historically, many leaders viewed HR systems as administrative plumbing. As this shifts to a strategic intelligence layer, what are the first concrete steps a company can take, and what new metrics can prove this transition is creating business value? Please provide a detailed example.
The first step is a mental shift, moving from seeing HRtech as a “system of record” to a “system of insight.” It begins with asking different questions. Instead of, “Are we compliant?” leaders need to start asking, “Which teams are at risk of burnout?” or “What skills will we be missing in six months?” Practically, this means connecting previously siloed data. A company could start by integrating its Learning Management System (LMS) with its performance management tool. Suddenly, you’re not just tracking course completions; you’re measuring the direct impact of training on performance scores and project outcomes. For example, a company might notice a team’s delivery speed has slowed. The old way was to wait for the quarterly review. The new way is to look at real-time collaboration data, see that communication has dropped, and cross-reference it with capacity analytics showing they’re overloaded. The metric for success isn’t just “training hours completed,” but “a 15% reduction in project delays” or “a 10% increase in productivity for teams who underwent specific skills training.” That’s how you prove the plumbing has become the intelligence layer—by connecting people data directly to financial and operational outcomes like revenue, margins, and delivery speed.
Companies often rely on static data like job titles and annual performance scores. How can an organization begin capturing dynamic, real-time signals like collaboration patterns and skills usage, and what cultural shifts are needed to make this data actionable? Please outline the first few steps.
Moving beyond static data feels like switching from a black-and-white photograph to a live video feed. The first step is to leverage the tools you already have but aren’t fully using for intelligence. Look at your collaboration platforms, your project management software, and your internal communication tools. These are goldmines of dynamic signals. You can start by analyzing metadata—not the content of messages, but the patterns of connection. Who is collaborating most frequently? Which teams are becoming isolated? This data shows how work actually gets done, far better than any org chart. The second step is to start inferring skills from actions. Instead of relying on a dusty resume, modern platforms can identify skills based on the projects an employee contributes to, the code they commit, or the internal documents they author.
The cultural shift is the harder part. There must be a move towards transparency and trust. Employees need to understand that this data is being used to support them—to identify burnout risks, find growth opportunities, or offer relevant training—not for surveillance. Leadership must be trained to interpret these signals as indicators of system health, not individual fault. For instance, if data shows a team is disengaging, the conversation shouldn’t be about blame. It should be about workload, resource allocation, or leadership support. The goal is to use these continuous signals to make work better, which requires a culture that sees data as a tool for proactive problem-solving, not retroactive punishment.
The concept of AI in HR is moving from simple automation to complex orchestration. Can you explain the practical difference between these two and share an anecdote of how AI orchestration helped a company predict a skill gap or reduce burnout risk before it became a crisis?
The difference between automation and orchestration is like the difference between a player piano and a symphony conductor. Automation makes a single process faster, like automatically scheduling interviews or processing payroll. It’s efficient but unintelligent. Orchestration, powered by AI, is about conducting the entire symphony of the workforce. It connects disparate signals from hiring, learning, and performance to make intelligent, coordinated decisions across the entire employee lifecycle. It doesn’t just speed up a task; it makes the entire system smarter.
I worked with a tech company that was constantly surprised by attrition in its data science division. They were automating parts of their exit interviews, but that was just reacting faster. By implementing an AI orchestration layer, we connected signals from their learning platform, collaboration tools, and project management system. The AI model noticed a pattern: top performers who were about to resign first showed a decline in their learning activity, followed by a withdrawal from cross-functional collaborations, and finally, a drop in their engagement in team channels. It flagged three high-value data scientists who fit this pattern. Instead of waiting for them to quit, their managers were alerted. They intervened, discovered the employees felt their skills were stagnating, and moved them onto a new, challenging AI project. The company didn’t just prevent three costly departures; it shifted from reacting to attrition to proactively managing talent and engagement. That’s orchestration.
With advanced analytics, HR leaders can evolve into “workforce economists” who partner with finance and operations. What does this role look like day-to-day, and how does it change the strategic conversations HR has with the CFO? Please walk us through a typical scenario.
A “workforce economist” thinks about talent the way a CFO thinks about capital. Their day isn’t about handling employee relations issues; it’s about analyzing the supply, demand, and ROI of human capability. Instead of looking at headcount reports, they’re modeling the financial impact of skill scarcity, the cost of attrition versus internal redeployment, and the productivity return on investments in development programs.
Let’s walk through a scenario. The CFO comes to a strategy meeting and says, “We need to cut operational costs by 10% to improve our margins.” The old HR response would be to discuss layoffs. The workforce economist, however, comes prepared with a different analysis. They present a dashboard showing that while two departments are overstaffed, three others are suffering from critical skill gaps that are creating project bottlenecks and costing the company money in delays and rework. They model a scenario: instead of layoffs, what if we invested in a targeted reskilling program to redeploy talent from the overstaffed areas to the understaffed ones? They can show the CFO the projected cost of the reskilling program versus the much higher cost of attrition, recruitment, and lost productivity from the skill gaps. Suddenly, the conversation with the CFO isn’t about cutting heads; it’s about reallocating human capital for a higher return. HR is no longer a cost center but a partner in driving financial performance.
Fragmented HR tools, or “tool sprawl,” can create more noise than insight. How does an “HRtech Operating System” solve this by standardizing data, and what is the most critical element for ensuring different platforms can successfully communicate with each other? Share a step-by-step approach.
“Tool sprawl” is a massive problem. You have one system for hiring, another for learning, a third for performance, and they all define “skill” or “team” differently. You get a dozen dashboards that tell conflicting stories. An HRtech Operating System (OS) solves this by acting as a unifying infrastructure, a common ground where all these tools can speak the same language. It’s not about replacing every tool, but about integrating them through a shared data layer.
The most critical element for communication is an API-driven architecture combined with a standardized data model. This means that every tool, whether it’s for recruiting or collaboration, plugs into the central OS and agrees on core definitions—what constitutes a skill, how performance is measured, how a team is defined.
Here’s a step-by-step approach to building this: First, conduct an audit of all your existing HR tools to identify overlaps and data inconsistencies. Second, define a “golden record” for your workforce data. Establish a single, authoritative definition for key entities like employee, role, skill, and team. This becomes your common language. Third, prioritize integration based on strategic value. Start by connecting the systems where the data flow is most critical, like linking your skills data from the learning platform to your project staffing needs in the performance system. Finally, insist that any new tool you purchase has robust, open APIs. The goal is to build an ecosystem where data flows continuously, turning fragmented noise into a clear, unified signal about the health and capability of your organization.
What is your forecast for the future of HRtech?
My forecast is that the line between HRtech and business strategy will completely disappear. We will stop talking about it as “HR technology” and start seeing it as the enterprise’s “organizational intelligence infrastructure.” The HRtech OS won’t just be a system for HR leaders; it will be a decision-making co-pilot for every executive, manager, and team leader in the company. AI will move beyond prediction to prescription, not just telling leaders who might leave, but suggesting a series of orchestrated actions—a development opportunity, a conversation with a mentor, a change in workload—to proactively address the root cause. This will redefine the role of managers, turning them from supervisors into coaches empowered with real-time insights. Ultimately, the future of HRtech is about moving from managing human resources to architecting human capability, creating organizations that are not just efficient, but truly adaptive, resilient, and intelligent from the ground up.