The Problem Nobody Talks About: Why Training Spend Isn’t the Same as Performance
The Question Nobody Wants to Answer
How much does your organization spend on training every year? Now the harder question: how much of that actually changes performance — not how much people liked it, not how much they learned, but what changed in what people do and what the organization gets as a result. On the debut episode of a new series unpacking the Kirkpatrick Model, Vanessa Milara Alzate argues that most organizations can’t answer that question, and that the gap is costing far more than anyone is tracking.
The Real Cost of the Gap
People costs are typically the single largest controllable line on a P&L — bigger than technology, real estate, often bigger than marketing. So when that investment doesn’t produce a return, it isn’t a footnote. Losing an employee costs 30% to over 200% of their annual salary once recruiting, onboarding, ramp-up time, and lost knowledge are counted. Employees who feel under-trained after onboarding are dramatically more likely to leave — a baseline turnover risk that nearly doubles when training falls short. Leadership development alone is a $60 billion industry, yet by some estimates only about 10% of that spending produces results anyone can point to. Layer in the cost of rework, escalations, customer churn, and safety incidents that happen when people were technically trained but never changed what they do, and the number gets large fast — even when nobody is calculating it.
Where the Kirkpatrick Model Actually Came From
The story starts in the 1950s, when Dr. Don Kirkpatrick was teaching a human relations program for industrial supervisors at the University of Wisconsin and, as part of his PhD dissertation, set out to answer a practical question: did this training actually work? In 1959, when ASTD (now ATD) asked him to write an article on his evaluation work, he proposed four — one each on reaction, learning, behavior, and results. He didn’t call them “levels,” and he didn’t draw the triangle graphic everyone associates with his name; the industry gave the model that shape later, in the 1970s. But the four questions were his: did people like it, did they learn something, are they using it, and is it making a difference?
The Four Levels, In Plain Language
Level one, reaction, asks whether people found it valuable — not just satisfaction, but engagement and relevance to the work ahead. Level two, learning, asks whether they learned what they needed: knowledge, skills, attitudes, confidence, and commitment to apply it. Level three, behavior, asks whether they’re doing what matters on the job, and whether the environment enables or blocks that new behavior. Level four, results, asks whether it made a difference, tracked through leading indicators along the way and lagging indicators at the end.
The Model Evolves: Behavior Is Where the Magic Happens
Decades later, Dr. Jim Kirkpatrick and Wendy Kirkpatrick Kayser built the New World Kirkpatrick Model, restoring teachings that had been forgotten and centering relevance at level one and confidence and commitment at level two. They emphasized something critical: level three contains the magic of the model, supported by required drivers — the specific on-the-job structures and accountability tools needed for behavior change to hold. The real shift was starting with the end in mind: begin with the result, then work backward to the behaviors, learning, and experiences that make it possible. Today’s model goes further still, recognizing that training doesn’t operate in isolation — it sits inside a performance environment of systems, leadership, culture, and resources that can support progress or quietly block it, no matter how well-designed the training is.
Why the Model Still Works
The model endures because it’s both simple and strategic — it gives everyone, not just learning teams, a common language to connect effort with outcomes, and it moves organizations from “train and hope” to designing for performance. Two deeper indicators sit underneath it: Return on Expectations (ROE), the ultimate measure of value delivered to the business, and Return on Performance (ROP), the ultimate measure of how well the initiative itself was executed — both explored later in the series.
Your Assignment This Week
Take a current initiative in your organization — not necessarily training. What would success actually look like? What result would tell you it worked, and what do people need to do differently to get there? Too often, a problem shows up — sales drop, errors increase, customers complain — and the default response is “we need training,” when that may not be the real problem at all. The next episode in this series tackles that head-on.
It’s back-to-school season, which makes this a natural moment to pick up Building a Culture of Evaluation and follow along with this series. Vanessa will host a live Book Club discussion with readers this fall — details, and the book
Join the book club here: https://lp.constantcontactpages.com/sl/lSZFbHe
Grab a copy of the book here: https://www.amazon.com/Building-Culture-Evaluation-Kirkpatrick-Performance/dp/1963392337/
