The conversation usually starts around slide 14. The CFO is nodding but not listening. Board members are cycling through their phones in that particular way that signals the decision was already made two slides ago. The CHRO is explaining, for the third time in as many quarters, why the learning and development budget should be protected. They have satisfaction scores. They have completion rates. They have a testimonial from a senior leader who called the last programme "genuinely transformative." The CFO asks what the programme costs. The CHRO says the number. The CFO asks what it produces in terms the P&L recognizes. The CHRO reaches for another slide.
This is not a budget conversation. That is the first error, and it is the CHRO's error to own. The CHRO is defending a line item when the conversation they need to be having is about governance. The distinction sounds like semantics until you understand that every organization has a small number of structural mechanisms that determine how decisions get made, how people advance through the hierarchy, which behaviors replicate and which get quietly eliminated. L&D is one of those mechanisms. Most CHROs have never positioned it as one. They built it as a service function, staffed it like a service function, and then discovered that services get cut when the numbers get hard.
How L&D Became a Cost Center by Design.
Trace it backwards. L&D functions in most Fortune 500 organizations were constructed as training delivery units. Their mandate was to run programmes. Programmes were measured by participation. Participation became the evidence of value. Over twenty years, the infrastructure of the function built itself around producing evidence of participation rather than evidence of performance. Reporting lines went into HR. Budgets were allocated per headcount, not against business outcomes. Vendors were contracted to deliver content, not to architect the systems that change how an organization behaves at scale. The function became, in the most precise description available, a museum of completed activities.
Three specific design choices made this permanent rather than accidental.
The first choice was measurement language. When L&D decided to report on completion rates and satisfaction scores, it chose a vocabulary the business does not speak. Finance thinks in terms of performance output. L&D built its evidence base in terms of process input. These are permanently misaligned frames. The CFO did not impose this misalignment. L&D built it, maintained it, and then defended it as rigorous for two decades running.
The second choice was governance placement. A function that reports into HR, which reports to a CHRO who is several steps removed from the CEO's strategic conversations, does not have the organizational standing to make a board-level argument. The reporting structure communicates the priority before the first slide appears on screen. The room has already categorized the conversation before anyone speaks.
The third choice was vendor architecture. When organizations outsource content delivery to vendors, they are renting exposure to ideas. They are not building capability. The vendor leaves after the programme ends. The architecture, the reinforcement systems, the judgment-development infrastructure that was supposed to remain. Except it was never built. What remains is a list of programmes people completed, a stack of feedback forms rated between four and five stars, and no lasting change to how the organization makes decisions when the pressure is real.
The function was built to produce evidence of activity. The board is asking for evidence of performance. These are not the same question, and no amount of additional reporting on the wrong thing will close that gap.
The Narrative and What It Costs.
The narrative the organization tells itself continues in parallel. "We invest in our people." This line appears in annual reports, ESG filings, earnings calls, and town hall presentations. It is not a lie. The organization does invest in people. It invests in their exposure to content. What it almost never invests in is the architecture that converts that exposure into behavioral change at scale. The investment and the outcome are treated as equivalent in public communications. Nobody corrects this in a board meeting because the people who would need to correct it are the same people who approved the investments.
Here is what no one is measuring.
Not satisfaction scores. Not completion rates. Not NPS from the last cohort.
The unmeasured cost is the quality of decisions made by the leaders the L&D function was designed to develop. Every leader who makes a poor judgment call under ambiguity. Every escalation that should never have left the floor it started on. Every missed market signal. Every quarter where the organization moved too slowly because the leadership layer three levels below the executive team was waiting to be told what to do, because they had always been told what to do, because the system built to develop their judgment had delivered a strategic thinking course they completed in 47 minutes and rated four stars out of five.
These costs are real. They are large. They do not appear on the L&D dashboard.
The Response That Makes It Worse.
When the board gets skeptical, most L&D functions respond by adding more measurement. More surveys. Before-and-after knowledge assessments. More data-gathering infrastructure to demonstrate impact. This is the wrong move, and it compounds the underlying problem. The issue was never a lack of data. The issue is that the data the function collects does not answer the question the business is asking. Adding more instruments to measure the wrong thing produces more evidence that the function is misaligned with what the organization needs. The board does not become more confident. They become more specific in their skepticism.
Some CHROs pivot to technology. A new learning management system. An AI-powered content platform. A digital library with fourteen thousand courses available on demand. This is the most expensive version of the same mistake. Technology that delivers content faster is not architecture. A more efficient museum is still a museum. The fourteen thousand courses will be ignored at roughly the same rate as the two thousand courses they replaced. The utilisation data will come back underwhelming. The board will ask why the new platform is not being used. Nobody will have a satisfying answer, because the fundamental design problem has not changed.
The CHRO's hidden leverage has never been the budget. It has been the architecture.
Every large organization is, in operational terms, a machine for replicating judgment patterns. The decisions that the top layer makes get encoded into processes, precedents, and expectations. The layer below learns from those patterns. The layer below that inherits the result. Whoever designs the system through which future leaders are developed controls which judgment patterns get replicated into the organization's next decade. That is not a human resources function. That is a governance mechanism. One with a direct and measurable line to organizational performance.
The CHRO is sitting inside one of the most powerful leverage points in the organization. They are presenting it as a cost center because they designed it as one. The CFO is not stripping authority from the CHRO in that boardroom. The CHRO assigned themselves the wrong role years before the meeting was scheduled.
What Rebuilding This Actually Looks Like.
When L&D is repositioned as a performance governance mechanism, the design questions change entirely.
Instead of "what content do leaders need," the question becomes: what judgment patterns does this organization need in order to make the decisions the next three years will require? What architecture builds those patterns at scale? What measurement system tracks them over time, across the layers where those patterns actually matter?
Instead of participation rates, the function tracks decision quality metrics. Escalation frequency. Response accuracy under ambiguity. How quickly poor judgment compounds before it surfaces at a level where it becomes visible. How long it takes for a decision made at the wrong level to reach the right one. These numbers exist in every organization. Nobody is currently collecting them under the L&D umbrella, because the function was never designed to ask for them.
Instead of vendors who deliver programmes, the architecture requires people who design systems that persist after the engagement ends. The difference between a content delivery vendor and a performance architect is the same as the difference between a caterer and a nutritionist. One shows up, provides what was ordered, and leaves. The other changes how the organization feeds itself going forward.
The budget conversation with the board changes because the evidence changes. Not "here is what we spent and how many people attended." Instead: here is the performance architecture we have built. Here is how it changed how this organization behaves under pressure. Here is the decision quality data, mapped against business outcomes, that demonstrates the function is producing what the organization needs it to produce. That is a conversation the CFO can engage with. It speaks in the language of organizational performance, not programme participation.
At SSUNDAR, the diagnostic work begins by mapping what the current L&D architecture actually produces versus what the organization believes it produces. The gap between those two assessments is, without exception, larger than the CHRO anticipated. The rebuild that follows is not a programme redesign. It is an architecture redesign. One that gives the CHRO something the board can evaluate on terms the business actually cares about. The kind of evidence that makes slide 14 unnecessary, because the performance data is already in the room before the meeting starts.
The CFO is not the enemy. The measurement system the CHRO built is.