What should we measure to tell whether our training worked?
To tell whether your training worked, measure four things in order: how employees responded to it, what they learned, whether they use it at work, and whether the outcome the training was meant to improve has changed. Each one tells you something the others can't, and the last two are the most important when you're judging whether the training was worth its cost.
| What to look at | The question it answers | Where the information can come from |
|---|
| Response | Did employees find it relevant, useful, and engaging? | A few questions or a short conversation right after the training |
| Learning | Do they know something, or can they do something, they couldn't before? | Practice exercises, demonstrations, or short checks during or after the training |
| Use on the job | Are they doing things differently at work? | Supervisors, coworkers, the employees themselves, and observation, a few weeks or months later |
| Results | Did the problem the training was meant to address improve? | Records you already keep, such as incidents, errors, rework, turnover, or time until new employees work on their own |
You don't need to measure all four with the same effort. For a short, inexpensive training, a few questions and a conversation with supervisors may be enough, while for a larger or ongoing investment it's worth planning the measurement before the training begins. That way you have a starting point to compare against, and the information you collect is information your organization will use.
What if the training was never tied to a specific problem?
If the training was never tied to a specific problem, go back and find out what prompted it, because you can't judge whether something worked without knowing what it was supposed to accomplish. Organizations often offer training because it seems like good practice, because a manager requested it, or because something felt wrong without anyone identifying exactly what.
Ask the people who requested or approved the training what they hoped would be different afterward, and then ask employees and supervisors what they see as the most significant gaps in skills or knowledge. Sometimes you'll find that the training addressed a real need. In other cases you'll find the problem was something training can't fix, such as unclear expectations, inconsistent supervision, an unmanageable workload, or a process that makes the work harder than it needs to be. Knowing which of these you're dealing with is worth finding out before you spend more on the next round.
How do we weigh what training costs against what it's worth?
To weigh what training costs against what it's worth, compare the full cost of the training with the cost of the problem it was meant to address, and be honest about how much of any improvement the training can account for. The full cost includes fees, materials, and travel, plus the hours employees spend in training and the work that isn't done while they're there. The cost of the problem might be the turnover, errors, rework, safety incidents, or overtime the training was meant to reduce. What is turnover actually costing us? shows how to estimate one of those costs from your own records.
Be careful about attributing every improvement that followed to the training itself. A new supervisor, a slower season, or a change in hiring practices can affect the same outcomes. Comparing employees who took the training with similar employees who haven't yet, or comparing the same group before and after, can help you judge how much of the change to attribute to the training. How do we know whether the change we made actually helped? covers those comparisons in more detail.
Why didn't the training change how people work?
Training may not change how people work for several reasons, and it helps to find out which one applies before concluding that the training itself was poorly designed. Employees may not have had an opportunity to use what they learned soon enough to remember it, or their supervisors may not know what the training covered or may not support the new approach. The way the work is organized may make it difficult to do things differently, and in some cases the training simply didn't connect to the situations employees encounter in their jobs.
Check how the training was carried out, too: who attended, whether they attended all of it, and whether it was delivered the way it was planned. If results fall short, that doesn't automatically mean the underlying idea was ineffective. If employees found the training irrelevant or a poor use of their time, Why do employees roll their eyes at our training programs? looks at why that happens and what to do about it.
The Grantibly perspective
Lauren's perspective
When a leader asks us whether a training was worth it, the first thing we want to know is what it was supposed to improve. Instead of asking whether employees liked the changes, we want leaders to ask whether the changes improved the outcomes they were trying to improve.
We don't believe organizations need more data simply for the sake of having more data. We would rather see a focused look at what changed, using records the organization already keeps and the observations of the people closest to the work, than an enormous measurement plan that creates work for everyone and produces information no one has time to use.
We also avoid making unsupported promises about return on investment. No one can responsibly tell you in advance exactly what a training will return. What good information can do is help you determine where an investment is likely to make a meaningful difference, and where it isn't.
