The Two-Word Test That Saves Training Budgets

A manufacturing plant I worked with had a quality problem. Defect rates on one production line had crept up for three straight months, and the plant manager wanted retraining for the line operators. Standard procedure, standard fix, standard budget request already halfway approved by the time I got involved.

Before signing off on anything, I asked the plant manager a single question. If you offered each operator on that line one lakh (100 thousand) rupees to run one unit through the process perfectly, with a supervisor watching, do you think they could do it?

He didn't hesitate. Of course they could. These were experienced operators, some with a decade on that exact line. They knew the process inside out.

That answer ended the conversation about training before it started. If people can do something correctly under the right conditions, teaching them how to do it again isn't going to help. Something else was going wrong, and it had nothing to do with what was in anyone's head.

Workers in a bustling factory diligently operate machinery, with sparks flying as they skillfully shape metal components, capturing the essence of industrial craftsmanship.

The Test Itself

The question I asked him has a name in performance improvement circles, and it comes from a slim, decades-old book by Robert Mager and Peter Pipe called Analyzing Performance Problems. The test boils down to two words: can't versus won't.

Here's the full version. If someone's life depended on performing the task correctly, right now, with no advance notice, could they do it? If the honest answer is yes, you're looking at a won't-do problem. Something is blocking the behavior that has nothing to do with ability: motivation, incentives, unclear expectations, a broken process, competing priorities, fear, habit, or plain old inertia. If the honest answer is genuinely no, even under that kind of pressure, then you might actually have a can't-do problem, and training becomes a reasonable candidate for fixing it.

That's the whole test. Two words, one question, and an uncomfortable amount of honesty required to answer it properly.

Why It Works So Well

The reason this test is so useful isn't that it's clever. It's that it's fast, and it forces a kind of clarity most performance conversations never reach. Ask someone directly, "could this person do it if their life depended on it," and you strip away all the polite ambiguity that usually surrounds a performance problem. Nobody can hide behind "well, it's complicated" when the question is framed that starkly.

I've watched this test cut through hours of circular debate in a single exchange. A stakeholder insists the team needs training. You ask the can't-do question. The room goes quiet for a second, because everyone already knows the answer, they just hadn't said it out loud yet. Once somebody admits "yeah, they could do it, they just don't," the entire conversation shifts. You stop talking about curriculum and start talking about what's actually stopping people.

There's a deeper reason it works too. Most organizations default to training as a solution because training is visible, budgetable, and safe to propose in a meeting. Nobody gets criticized for suggesting a workshop. But the can't-do test reroutes that default by making the cost of guessing wrong explicit and immediate, right there in the room, before a single dollar gets spent. It replaces a vague hope that training will help with a specific, falsifiable claim: this person cannot do this, full stop, under any conditions. That claim is either true or it isn't, and you can usually find out within a day.

What a Won't-Do Problem Actually Looks Like

I want to spend a minute on won't-do problems specifically, because they get misdiagnosed as training gaps constantly, and the misdiagnosis is almost always more expensive than the original problem.

Take a customer service team where agents technically know the escalation process but rarely use it, choosing instead to let frustrated customers sit in a queue longer than they should. Ask the can't-do question and the answer is obviously yes, they could escalate correctly if they wanted to. So why don't they? In one case I looked at, the answer turned out to be depressingly simple: escalating a call made an agent's average handle time look worse on the dashboard their manager reviewed every week, and handle time was the metric tied to their quarterly bonus. Nobody had built training around escalation because nobody needed it. What they'd built, accidentally, was an incentive structure that punished the correct behavior.

No course fixes that. You could run agents through the escalation process a hundred times and the bonus structure would still be quietly telling them not to use it. The fix here wasn't instructional at all. It was a five-minute conversation with the operations team about adjusting one metric.

I've seen versions of this story play out with sales teams told to "prioritize long-term relationships" while getting measured purely on this quarter's closed revenue, with warehouse staff told to "follow safety protocol" while informal culture rewarded speed over caution, and with managers told to "coach their people more" while their calendars left no actual time in the week to do it. In every case, the people involved could do the desired behavior. They'd proven it before, under the right conditions. What they lacked wasn't skill. It was a reason, or the room, or the permission.

What a Genuine Can't-Do Problem Looks Like

To be fair to training as a solution, real can't-do problems absolutely exist, and the test is just as useful for confirming them as it is for ruling them out.

A software company I worked with rolled out a completely new billing platform, and support agents started fumbling basic customer questions almost immediately. Ask the can't-do question here and the answer was a genuine no. These weren't experienced people avoiding something they already knew how to do. The platform was brand new, nobody had used it before, and there was no amount of motivation or incentive redesign that was going to conjure familiarity with an interface people had literally never touched. That's a can't-do problem in its purest form, and it called for exactly what everyone's first instinct suggested: structured training on the new system, ideally with hands-on practice before go-live rather than after.

The test isn't biased against training. It's biased against guessing. Sometimes the guess turns out right, and the test just confirms it quickly instead of after a wasted quarter.

Using It in Your Next Meeting

If you want to try this the next time someone brings you a training request, here's the version I suggest, because "if their life depended on it" can sound a little dramatic in a Tuesday afternoon status meeting. I usually ask it like this: "If you paid this person a large cash bonus to do it correctly, right now, with no prep time, do you think they could?"

Watch how people answer. A confident, immediate "yes" is your signal to stop talking about training and start asking what's actually blocking the behavior. A genuine, considered "no, I don't think they could" is your signal that a learning intervention deserves serious consideration. And a hesitant "well... probably, but..." is often the most useful answer of all, because that hesitation is usually pointing straight at the real problem, if you're willing to follow it.

The test costs nothing to run. It takes fifteen seconds to ask and rarely more than a few minutes to answer honestly. Compare that to the cost of building, delivering, and evaluating a training program that was never going to touch the actual problem, and the math isn't close.

A Small Habit With a Large Payoff

None of this requires new certifications or a different job title. It requires building the habit of asking one blunt question before agreeing to build anything, and then having the nerve to sit with whatever answer comes back, even when that answer means telling a stakeholder that the fix they wanted isn't the fix they need.

I've used this test on manufacturing floors, in sales organizations, in hospital systems, and in customer service centers, and it has never once failed to sharpen the conversation. Sometimes it confirms that training really is the answer, and that confirmation is worth having too. Most of the time, though, it saves a budget, a quarter, and a team's patience by catching a wrong diagnosis before it turns into an expensive one.

Two words. One question. Try it in your next meeting before the training request gets approved, not after.