The Pattern Nobody Talks About
A leadership program launches with a kickoff event, a cohort of excited participants, and a line item in the budget that got approved without much debate. Eighteen months later, nobody can say what it changed. Attendance drops. The budget gets flagged in a cost review. The program ends quietly, and the explanation that circulates is always some version of “it cost too much for what it delivered.”
That explanation is usually wrong. Most talent programs that get cut were never expensive relative to the problems they were meant to solve. They get cut because nobody connected them to a number the business already cares about. A program with no link to a business outcome is a program with no defense when budgets tighten, regardless of what it costs.
Start With the Business Question, Not the Program Idea
The programs that survive tend to start from a different place. Instead of “we should build a mentorship program” or “we need more training,” the starting question is something like: why are we losing people in their second year, or why does it take so long to fill a director role internally when we have qualified people three levels down.
That question points to a metric. The program becomes the answer to the metric, not a standalone initiative that gets evaluated on vibes and survey scores. This is the difference between a program that has a sponsor who will fight for it in a budget meeting and one that has an owner who hopes nobody asks hard questions.
Four Numbers Worth Wiring a Program To
Most people functions already have the data for these, even if nobody has pulled them together.
Regrettable attrition. Not total turnover, which includes people the business was fine losing. The number that matters is how many high performers in critical roles left in a given period, and whether that number moved after a program launched.
Internal mobility rate. The share of open roles, especially senior ones, filled by people already inside the organization. A low number here is often the clearest sign that development programs exist on paper but are not producing people who are actually ready when a seat opens.
Time to productivity. How long it takes a new hire or newly promoted person to perform at the level the role requires. Programs aimed at onboarding or ramp-up should be able to show this number shrinking.
Bench strength for critical roles. How many people are identified as ready now, or ready in one to two years, for the roles that would hurt the most to leave unfilled. This number is harder to track than the others, but it is the one that gets a CFO’s attention fastest, because it maps directly to business risk.
What a Finance Leader Actually Needs to See
A CFO does not need to understand the design of a leadership program. They need to see that a dollar spent on it moved a number connected to cost or risk. A simple model works better than a sophisticated one: here is the metric, here is where it stood before, here is where it stands now, here is what it would cost to solve this problem by hiring externally instead. Drew Soule, a Senior HR Business Partner and Organizational Design Consultant, has pointed to this kind of translation work as the real test of whether a people function is operating as a business partner or as a service desk that happens to run programs.
Spotting the Programs That Are Just Theater
Not every program needs to survive, and some honesty here saves a lot of wasted effort. A program is likely theater if nobody can name the metric it was built to move, if the only evidence of success is attendance or satisfaction scores, or if it has run for more than a year without anyone asking what changed because of it. None of that means the people running it were careless. It usually means the program was approved on instinct, which is an easy thing to do when budgets are healthy and nobody is asking yet.
Building the Habit Early
The organizations that get this right do not wait until a downturn forces the question. They build the habit of wiring every program to a number before it launches, while things are going well and there is no pressure to justify the spend. Drew Soule‘s view, shaped by work across engineering, manufacturing, and fast-growing technology environments, is that this habit is cheaper to build early than to retrofit under pressure. By the time a budget review forces the question, it is often too late to produce the data that would have made the case. The fix is not a bigger budget or a flashier program. It is making sure, from the start, that someone could answer the question a finance leader will eventually ask: what did this change, and how do we know?
