They were wrong. We were wrong. Executives and managers are about the business of satisfying multiple stakeholders by securing results. So, simply suggesting that training is an end in and of itself is both simplistic and insensitive. In the corporate world, if you don’t satisfy stakeholders you’re out of business.

This being the case, training designers have to make the following connection. You train people. Then, in addition to thinking or feeling differently, these newly trained people actually act differently. These changes in behavior, in turn, lead to changes in results. Better yet, they lead to changes in the results your company is trying to improve at the moment. Anything short of this and you’re merely entertaining, building teams, creating trust, or doing whatever suits you. However, what you’re not doing is improving key corporate results and well, you ought to pack up your tent and move on.

With this in mind, all interpersonal skills training needs to be created in the following fashion.

First, identify the key business result you’re trying to improve. Never create or deliver training without aiming it at a key performance indicator. To find the correct indicator, ask yourself what you really want. Not all indicators are equal. In some industries, people hardly ever ask what they really want. For instance, with NGOs (charitable institutions), managers routinely aim their efforts (including their training) at getting their product into the hands of the poor. They figure that the more people they hand out wheat to, the better.

Nobel Laureate Muhammad Yunus turned this strategy on end when he pointed out that the primary goal of many NGOs should not be to cure daily hunger pangs but to lift the poor permanently out of poverty by creating jobs that generate more than a dollar per day. Only by asking what they really wanted, was Dr. Yunus able to re-direct the efforts of thousands of NGOs toward the better goal of self-sufficiency and away from merely handing out resources. So, look at your key performance indicators, ask what you really want, and then select from or create an indicator that measures what you value. This is the result you’ll try to affect through training.

Next, identify the key behaviors that, if changed, will improve the result you want. Herein lies the rub. It’s hard to know what people need to do differently in order to achieve the result they want. Hard or not, trainers have to find the link or they won’t know where to aim their efforts. That’s how generic communication training or leadership courses are often created. Trainers are not quite sure what people need to do differently, but they figure if they get better at communicating well, it can’t hurt. Right? Wrong. If the actions you’re teaching are not directly and clearly linked to key results, the training is likely to take time, cost money, fail to get results and thus undermine the credibility of the training and the training facilitators.

Conduct a positive deviance study. To discover the specific behaviors that directly impact the results you want to achieve find other work groups, departments, or companies who are succeeding where you are routinely failing. Discover what these successful folks do that separates them from you and the rest of the pack. This is the practice of uncovering positive deviants.

For instance, when trying to uncover why one company had a near-perfect safety record in an industry that suffered many accidents, researchers discovered that the safe company didn’t have a safety program. Everyone else did. They didn’t hang banners or give speeches. Everyone else did. What they did do is speak up the first moment one of the researchers stepped into a hard-hat area without a hard hat. Several people spoke up—directly and politely. In the places that hung banners, people didn’t speak up. They thought it was rude or inappropriate. They let the banners do the talking and the banners didn’t work. Find the distinct behavior of successful people—in this case to speak up when people violate safety standards—and teach them to your trainees.

Copy others’ work. Commonly, once you’ve identified the measure you want to improve, the research that connects behavior to outcome has already been done for you. Scholars and practitioners have found the link between behaviors and results, and it’s your job to search them out and then stand on their shoulders. Whether you conduct the research yourself or rely on others, the point is the same. Don’t design training until you know the behaviors you want to change and don’t select the behaviors until you know how they affect the result you care about. Once again, if you can’t link the behavior to the result, you’re taking a huge risk.

Next, design training that affects the key behavior. Now you’re to the point where you have to design training that actually affects how people behave at work. Even if you’ve identified the right behaviors, if you can’t enable and motivate people to routinely enact the behaviors at work, training participants can’t or won’t do what’s required and your results will remain unchanged. While there are dozens of places to start when trying to motivate and enable new behaviors, here are several principles you should keep in mind if you want to see people actually behave differently once the training is completed.

Never confuse behaviors with outcomes. Many training courses offer the following type of advice: “When first bringing up a problem with someone, establish a good relationship.” This advice sounds good, but upon further reflection, it explains what to achieve, not what to do. Participants are being told to do something (not clear what exactly) that leads to a good relationship (an outcome). The outcome is clear, the action is unknown. On the other hand, behavioral advice informs action. It is recognizable (people can pick it out of a behavioral lineup) and replicable (they can also do it). Trainers often offer outcome advice because they haven’t followed the previous steps and don’t know the requisite behaviors.

Focus on both motivation and ability. Most training spends far too much time convincing people they ought to give up their old ways, and far too little time teaching and practicing new ways. For instance, HR professionals often use thorough 360^. survey techniques to convince people they need to change. Ironically, most participants can be motivated to change with a quick review of ineffective interpersonal techniques (often captured on video) and why they don’t work—followed by replacement behaviors. At this point, trainers should quickly move to the hows and whys of the replacement behaviors. The key is to work on ability far more than you work on motivation.

Employ deliberate practice to improve ability. Nobody would dream of competing in figure skating without having first been carefully coached. You’d want to be shown what to do, given a chance to do it, be observed by professionals, and then given advice on what to change and how to change it. You’d repeat this process over and over again until you got it right.

The same sort of deliberate practice should be used with leadership and other interpersonal skills training. The ratio of lecture and discussion to coaching and feedback must tip toward practice and away from theoretical musings if you expect people to master complex interpersonal skills. When you know the exact behaviors required to secure results, when you can demonstrate them clearly on video or through live modeling, and when people can see the link between those specific behaviors and key results—they’ll willingly practice them with feedback.

Rely on theoretical paths. You can’t tell people to follow rote steps and expect the steps to work in complex human interaction. So, when teaching interpersonal scripts, build a variety of possible responses from the other person into your theory. For instance, after bringing up a problem to another person, that person may not be able to do what’s required, they may not want to do what’s required, or both. You have to build these contingencies (through pathing) into your interaction model. You also have to buld in the skills to deal with each contingency. If not, once the conversation starts to unfold, it could easily follow a path trainees aren’t prepared to deal with.

Identify entry conditions. As you teach specific skills, take the time to identify when and why the skill is called for. Demonstrate the entry condition, talk about the entry condition, warn people that they’ll need to watch for the entry condition, and then build these conditions into the practice. For instance, when people aren’t motivated they say such things as “Who cares,” What’s the big deal?” or “I have higher priorities.” These types of statements are entry conditions. Once trainees recognize the entry conditions, they can apply their newly learned skills to motivate the person.

Test for efficacy. Once you’ve trained people on what to do and how and when to do it, the final step is to measure their ability to do what’s required. Provide trainees with interpersonal tests, measure their effectiveness, give them feedback, allow them to make corrections, and don’t move on until people have mastered the requisite behaviors.

Look for results. If you’ve done your homework, you can now turn your attention to the results you’re measuring. If people can and do enact the behaviors you’ve taught, if the behaviors, in turn lead to results, and if the results are the ones you truly want—you’ve designed the perfect training course. You haven’t trained because it’s something you do, because you happen to have a course catalogue, and because you’ve got a nifty room with cool gadgets. Rather, you’ve trained with the results in mind and your carefully executed training initiative has helped you to improve key metrics.