The art of good coaching. (Chief Concern).

Most CEOs know that people are a company's most important asset. But if you really believe that--and I do--then you must make a genuine effort to be a good coach.

Good coaching is about trying to impart experience. You try to point out the best way of doing something, not because you're so smart, but because you've seen it 100 times. If I can help people find a better way, then I've contributed to their overall success, and that's what coaching is about. Conversely, if your people don't get any better, it's your fault, too.

I coach my direct reports, the people in crucial positions in the company. I coach business teams when I do business reviews of a strategic or operating plan. I coach when I sit down to do personnel appraisals, though that's a smaller percentage of the time. Most of the time, I coach when I'm spending time with groups of people.

It's best to find opportunities to coach in real-time business situations as opposed to offline. If I hear a team predict a difficult first year as they get their act together in launching a new product, but they're convinced the second year will go off the charts, I say, "Wait a minute. That's not how it happens. I've been there too many times. You'll make some progress in the second year, but not anywhere near this dimension. You've got to think about it a different way to have any credibility in the marketplace and with your people." And they usually do.

Every meeting is a coaching opportunity. Even if I know what's going to be said, I can still try to make it a learning experience by asking the right questions. I might say, "I was surprised to see the legislation coming out of Washington that affects this business. What do you think about that?" And I'll find out what they know about the external environment. They may know a lot about business, but need to think more externally.

Alternatively, if someone knows a lot about the regulatory issues, I'll acknowledge that, but will want to get the conversation focused on areas where he's not particularly adept. I'm not trying to trap people -- I don't think that's fair. But I want to make sure they're thinking. And the next time they come in, they'll spend more time on these other areas, and the business will be better represented because of it.

Whether it's a person, a business leader or a team, I look for behavior or performance, and then I try to make overall contributions that improve on that. We've had plenty of good operating people who had very little strategic sense. No one wants to admit that he or she has strategic shortfalls, because it sounds like you're intellectually deficient or that you lack conceptual skills. We had one individual who had a great rapport with his people, a wonderful hand on operation, and delivered on his commitments, but didn't have a good concept of where his business needed to go. I told him, "What you need to do is get someone with those conceptual skills. Work with that person to fill in this vacuum, and your overall performance will improve and your career will be better."

But you also need to ask yourself: Are you a gasbag who wants to hear yourself talk, or are you actually helping people? There's a big difference and you have to be sufficiently honest with yourself to make that distinction. You can tell by whether performance improves. If the business teams don't get better, you're not being effective. You're just talking, not coaching.

Coaching and performance are intricately tied. You can get short-term performance by flat in some situations, but I'm talking about sustainable performance. That means you have people who are continually being coached to deliver execution.

Time has changed things as well. The way you coached 10 years ago isn't the way you coach today. The old Vince Lombardi or Bobby Knight command-and-control model doesn't work anymore. People aren't comfortable in that environment; they want to have a say. You've got to tailor your approach so that people find something they can identify with.

This is also a two-way street. It's not just the coach telling the players what to do, but listening to how the players respond. If you look at the more vivid examples of good coaching, it's the person who has the best interpersonal skills, the best communication, who knows when to press and when to praise. The ability to read people's thoughts and let them participate in the overall goal is a big part of success.

Jack Welch was a good coach. I wasn't there, but I have a feeling Ralph Larsen was a good coach at Johnson & Johnson.

Bear in mind, too, that the way you coach today isn't necessarily the way you'll be coaching 10 years from now. Companies change; environments change. You have to be continually thinking about changing to adapt to what may well be a different place.

When I came back to Honeywell, I found an environment that was disappointed because the merger with GE hadn't happened. People were disillusioned because the company hadn't made any earnings projections over the past nine months. A lot of people had one foot out the door because they didn't think they'd have a job.

I went to all of our people and told them why I thought it was still a good place to be. Then I got together with the people who had major roles in leading the company and tried to assure them that we would find our way. Did I have any doubts? Of course. But that can never show. You have to be convincing. Once people believe that things can be fixed, that's the message they send through the organization. A little success built on a little success, people got their self-confidence back and then it was just a question of momentum.

At Honeywell, we measure people on their ability to develop other people, so they've got to be coaching too. Through comments from employees, you identify those who are doing it and those who aren't. Those who aren't should have limited careers.

I'm reminded of one young person, with high potential and an elite education, including business school, who had all of the characteristics of being one of tomorrow's leaders. We gave him a small venture to run and he prepared a very sophisticated business plan, but basically he was incapable of interfacing with people. It was really dreadful. He thought that because of where he'd been, everyone was going to nod when he said something. But he found out other people had their own views and weren't nodding, and he dealt with them in a reckless way. People didn't want to work with him. I called him in, and said, "Look, Joe, you've got so many things going for you, but if you can't resolve this, you can't be successful. When you lead a business, you need people more than they need you." Ultimately, he couldn't grasp why people didn't think he was the Holy Grail. So he left--and I heard that he left his next job and his next job.

With more obviously talented people, you want to be more difficult. I don't mean beat up on them--that doesn't do the organization or its self-confidence any good--but expect more. When they ask me why I'm so hard on them, I say, "Because you're better than the other people, and you won't get any better if you don't get over these humps."

Who coaches me? I have a lot of loudmouths working for me, so I get feedback. The more experienced you get, the more realistic you are about situations where you could have done better. I have to assess myself and judge whether I am progressing. The day you think you're a finished piece of work, you have started downhill.

People used to say, "If you show me the training, I'll show you the performance." But there are a number of companies that have good training but not good coaching, and that has impacted their performance. IBM had good training sessions in the 1980s, but they didn't have good business leaders because they didn't have good coaching. Lou Gerstner brought that in, and they've become a lot better.

At the end of the day, my success will depend on how well I have developed and coached people. I could be the best technologist or the most advanced financial wizard, but those skills are not going to be sufficient to get to the level of performance we need. But if I identify, develop and coach great people, that will be the single differentiating factor.

Larry Bossidy is the chairman of Morristown, N.J.-based Honeywell and former CEO of AlliedSignal. He is the co-author, with Ram Charan, of Execution: The Art of Getting Things Done. Send comments to [e-mail].