- Avoid assets. Assets are liabilities. Owning something is a last resort. Leasing is better, renting is better still, and finding a way to do just as well without the something is best of all. Take telecom. Why buy a phone system in a world with Skype and cell phones? All that wire you put in the walls isn't coming with you when you move.
- Failure is always an option. If you never fail, you don't try very hard. You need to try risky things, but be able to pick up the pieces and move on if they don't work. If you're afraid to fail, you don't belong in a position of of responsibility. A good strategist almost always has a Plan B in reserve in case Plan A doesn't work out. To devise a realistic Plan B, you have to consider the ways in which Plan A may not work out.
- Try to be the dumbest guy in the room. If you're the smartest guy in the room, you're probably in the wrong room.
8.29.2006
A few simple ideas
I try to use a few simple ideas over and over. This creative laziness saves time. This way, I don't have to think as hard about certain decisions, so I can focus on other decisions. Here are my simple ideas:
3.18.2006
Meetings
Your typical subordinate might have a weekly meeting with you, a department meeting, and a company meeting. They attend one birthday party per month, which should be combined with either the company or department meeting. That’s three one-hour meetings per week. They may have one or two other monthly or quarterly meetings. That’s not so bad. They may accumulate other meetings, but people should be encouraged to establish such meetings for a limited time. ("We’ll have this project meeting weekly until the major release.")
If people need to meet semi-regularly, they should schedule a weekly time for the meeting, and cancel the meeting if there is no need that week. An easy way to do this is to schedule the time, and presume the meeting is canceled unless the organizer of the meeting issues an agenda the day before.
Late-starting meetings are insidious. If meetings usually start late, then people have an incentive to come late, so they waste less time. This often descends into competition to deliberately arrive last, as a way to demonstrate your importance. ("You must wait, so my precious time is not wasted.") The result is a drastic productivity hit. A meeting of six people starting 10 minutes late costs the company an hour in lost productivity. What is the opportunity cost of that hour?
To control the cost of meetings, you need to put a stop to these games. Start meetings on time. Tell your people that you start meetings on time, and in fact all meetings in your department start on time. If a meeting is more than three minutes late, and people are missing, start anyway. If key people are not present, cancel the meeting and leave.
The above works for meetings composed of people within your reporting tree, in most corporate cultures. Also, it works if you explain why it’s important to start meetings on time.
Very occasionally, you will be late for a meeting. Apologize sincerely. One day, you will arrive late for a meeting to find that the meeting was canceled and everybody left. This is a great day.
You will spend more time in meetings than the typical employee. You may spend most of your time in meetings, or else in on-going email discussions, which are meetings, except slowed-down by the low bandwidth of email compared to face:face communication. At least some of the meetings will be with your peers and superiors, so you don’t have so much control over how these are run. But, over time, you can make suggestions for improvement. Throwing a fit in front of your boss over the late start of a meeting usually does not have the desired effect.
In meetings, pay attention to the speaker. Look at them, and listen to their words. Take notes, if you want. Don’t daydream, check email, make side conversations, or read. Don’t take phone calls, at least not very often. People take their cues from you. If you act like the meeting is unimportant, everyone will think so, and the meeting will be a waste of time.
At the end of the meeting, it’s sometimes useful to ask yourself, "what could I have done to improve the meeting?" This is especially useful for a kind of meeting you don’t have lots of experience with. You should do this periodically for 1:1s. Often, in retrospect, a meeting would have been improved if you had not said something. Since you are a senior person, your words carry weight. Even if there is no real damage from something you said, it can still easily derail a meeting.
If people need to meet semi-regularly, they should schedule a weekly time for the meeting, and cancel the meeting if there is no need that week. An easy way to do this is to schedule the time, and presume the meeting is canceled unless the organizer of the meeting issues an agenda the day before.
Late-starting meetings are insidious. If meetings usually start late, then people have an incentive to come late, so they waste less time. This often descends into competition to deliberately arrive last, as a way to demonstrate your importance. ("You must wait, so my precious time is not wasted.") The result is a drastic productivity hit. A meeting of six people starting 10 minutes late costs the company an hour in lost productivity. What is the opportunity cost of that hour?
To control the cost of meetings, you need to put a stop to these games. Start meetings on time. Tell your people that you start meetings on time, and in fact all meetings in your department start on time. If a meeting is more than three minutes late, and people are missing, start anyway. If key people are not present, cancel the meeting and leave.
The above works for meetings composed of people within your reporting tree, in most corporate cultures. Also, it works if you explain why it’s important to start meetings on time.
Very occasionally, you will be late for a meeting. Apologize sincerely. One day, you will arrive late for a meeting to find that the meeting was canceled and everybody left. This is a great day.
You will spend more time in meetings than the typical employee. You may spend most of your time in meetings, or else in on-going email discussions, which are meetings, except slowed-down by the low bandwidth of email compared to face:face communication. At least some of the meetings will be with your peers and superiors, so you don’t have so much control over how these are run. But, over time, you can make suggestions for improvement. Throwing a fit in front of your boss over the late start of a meeting usually does not have the desired effect.
In meetings, pay attention to the speaker. Look at them, and listen to their words. Take notes, if you want. Don’t daydream, check email, make side conversations, or read. Don’t take phone calls, at least not very often. People take their cues from you. If you act like the meeting is unimportant, everyone will think so, and the meeting will be a waste of time.
At the end of the meeting, it’s sometimes useful to ask yourself, "what could I have done to improve the meeting?" This is especially useful for a kind of meeting you don’t have lots of experience with. You should do this periodically for 1:1s. Often, in retrospect, a meeting would have been improved if you had not said something. Since you are a senior person, your words carry weight. Even if there is no real damage from something you said, it can still easily derail a meeting.
2.18.2006
One-on-One Meetings
You must meet with every one of your direct subordinates for an hour, once a week. I don’t see an alternative to this. If you have 14 direct reports, that’s 14 hours out of your week, but so be it. If you have 50 direct reports, quit your job now.
OK, maybe you don’t need to quit if you have 50 reports. You need to create another layer of management, even if it is unofficial. There are many names for such crypto-managers: team lead, project lead, focal point. Carefully pick one or two people at a time, give each a small group to start with, and support them. After a while, you will be meeting with project leads once a week, and they will be meeting with their people once a week. In time, the Team Leads will probably become actual managers, and you might get a fancier title too.
Back to one-on-ones. That name is confrontational, so I usually abbreviat it 1:1. Schedule them for an hour. Often, they will run short, but sometimes it takes people 20 minutes to get warmed up, and you don’t want to have to cut them off just when they’re getting to the Thing That Is Really Eating At Them because the half-hour is up. Hold all the calls you can during these meetings. If you take a call during a 1:1, the 1:1 will probably not he productive and you might as well not have bothered. A nice side-effect of this approach is that, if the 1:1 runs short, you have some time at the end of the hour when you are unlikely to be interrupted.
If you can’t meet with each of your people for an hour every week, meet with them for an hour every other week or in extreme cases an hour every third week. You should be able to meet with at least 15 people per week.
Some engineers in particular are hard to get talking. You can ask them questions that aren’t too personal. If you’re new, ask them how long they’ve been at the company, what groups they’ve been in, what they did before, and so on.
The main thing you do at this meeting is listen. Being heard makes people happy. You want your people to be happy. If you’re lucky, you might even think of something that will actually help your subordinate. But listen first. Avoid the temptation to jump in and give a bunch of advice. Let them finish their story.
Then, ask some questions. This serves two functions: first, it gives you more time to think of something useful to say. Also, people usually know the answers to their problems, but they want a reality check on the answer. Ask them what approaches they have considered for their problem. If you like any of their approaches, say so. If you see down sides, ask if they have considered them. You could then discuss how things might work out.
If you have some advice, and you think it’s pretty good, then go ahead and share it.
If you don’t have any concrete advice, you can tell a story. Try to think of something similar that happened to you and what you did about it. If your solution didn’t work well, be sure to mention that.
Often, a subordinate will have a problem, and they will need something from you to solve it. They will need a resource, permission to take an action, waiver of a rule, and so on. They may need for you to talk to someone in a different department, and find out what’s going on. They may need your protection. If you can’t give them what they want, tell them so now. They will get a frowny face, but they will be a lot more unhappy if you lead them to believe they have your help, and then you let them down. If you need to think about it, tell them you need to think about it.
Sometimes, the person will ask you to do something that should not be done. Just say so. "Aww, I don’t think that’s a good way to resolve that problem."
Sometimes, the person will come to you with something you are morally and/or legally required to act on. They may have discovered illegal or unethical activities. They may be experiencing sexual harassment or a hostile work environment. They may bring you a problem from outside of work that you are compelled to act on, based on the trust they placed in you. Don’t be a coward: do what needs to be done.
Talk about positive things in 1:1s too. If you ask somebody how things are going, and they say, "great," ask why. Don’t dig around for something negative just so you’ll have a problem to solve. Sometimes, things really are going great and there are no pressing problems.
Take notes at your 1:1s. Review them at the next 1:1, if appropriate. It’s OK to review them in front of your subordinate. They know you’re busy. That you took notes respects them almost as much as if the meeting were so important to you, you were sure you’d remember every detail and no notes are necessary.
OK, maybe you don’t need to quit if you have 50 reports. You need to create another layer of management, even if it is unofficial. There are many names for such crypto-managers: team lead, project lead, focal point. Carefully pick one or two people at a time, give each a small group to start with, and support them. After a while, you will be meeting with project leads once a week, and they will be meeting with their people once a week. In time, the Team Leads will probably become actual managers, and you might get a fancier title too.
Back to one-on-ones. That name is confrontational, so I usually abbreviat it 1:1. Schedule them for an hour. Often, they will run short, but sometimes it takes people 20 minutes to get warmed up, and you don’t want to have to cut them off just when they’re getting to the Thing That Is Really Eating At Them because the half-hour is up. Hold all the calls you can during these meetings. If you take a call during a 1:1, the 1:1 will probably not he productive and you might as well not have bothered. A nice side-effect of this approach is that, if the 1:1 runs short, you have some time at the end of the hour when you are unlikely to be interrupted.
If you can’t meet with each of your people for an hour every week, meet with them for an hour every other week or in extreme cases an hour every third week. You should be able to meet with at least 15 people per week.
Some engineers in particular are hard to get talking. You can ask them questions that aren’t too personal. If you’re new, ask them how long they’ve been at the company, what groups they’ve been in, what they did before, and so on.
The main thing you do at this meeting is listen. Being heard makes people happy. You want your people to be happy. If you’re lucky, you might even think of something that will actually help your subordinate. But listen first. Avoid the temptation to jump in and give a bunch of advice. Let them finish their story.
Then, ask some questions. This serves two functions: first, it gives you more time to think of something useful to say. Also, people usually know the answers to their problems, but they want a reality check on the answer. Ask them what approaches they have considered for their problem. If you like any of their approaches, say so. If you see down sides, ask if they have considered them. You could then discuss how things might work out.
If you have some advice, and you think it’s pretty good, then go ahead and share it.
If you don’t have any concrete advice, you can tell a story. Try to think of something similar that happened to you and what you did about it. If your solution didn’t work well, be sure to mention that.
Often, a subordinate will have a problem, and they will need something from you to solve it. They will need a resource, permission to take an action, waiver of a rule, and so on. They may need for you to talk to someone in a different department, and find out what’s going on. They may need your protection. If you can’t give them what they want, tell them so now. They will get a frowny face, but they will be a lot more unhappy if you lead them to believe they have your help, and then you let them down. If you need to think about it, tell them you need to think about it.
Sometimes, the person will ask you to do something that should not be done. Just say so. "Aww, I don’t think that’s a good way to resolve that problem."
Sometimes, the person will come to you with something you are morally and/or legally required to act on. They may have discovered illegal or unethical activities. They may be experiencing sexual harassment or a hostile work environment. They may bring you a problem from outside of work that you are compelled to act on, based on the trust they placed in you. Don’t be a coward: do what needs to be done.
Talk about positive things in 1:1s too. If you ask somebody how things are going, and they say, "great," ask why. Don’t dig around for something negative just so you’ll have a problem to solve. Sometimes, things really are going great and there are no pressing problems.
Take notes at your 1:1s. Review them at the next 1:1, if appropriate. It’s OK to review them in front of your subordinate. They know you’re busy. That you took notes respects them almost as much as if the meeting were so important to you, you were sure you’d remember every detail and no notes are necessary.
2.04.2006
Leadership
Leadership is:
You need all three to be much of a leader.
You can be a good leader and a bad manager. If you can’t be a good manager, you’ll have to be a good leader. If you can’t be a good leader, you’ll have to be an awfully good manager.
Leadership is important, because with leadership, you can get a bunch of people to work towards a common objective. Even if the common objective isn’t well-chosen, this situation will turn out better than everyone pursuing a different goal. So leadership ability is heavily rewarded in business.
To an extent, leadership is trainable. Courage may not be trainable, but you can at least learn to use all the courage you have. With practice, you can improve your ability to convey your excitement somewhat. With practice, you will get much better at explaining why something is important.
Leadership is not the same thing as charisma. Charisma is the knack for making people like and support you, in the absence of reasons for doing so. Leadership is the knack for making people like and support the plan, usually with the aid of enthusiastic yet careful explanation of the reasons for doing so. Many people think they can’t become good leaders because they aren’t charismatic. But charisma is over-rated. Many leaders have little charisma, and there is some evidence that charisma is actually detrimental to management ability. In any case, getting people to like you isn’t nearly as important as getting them to enthusiastically do what must be done. Charismatic leadership is indispensable to cults, but it’s optional otherwise, and often not the best option.
- courage in the face of frightening things
- the ability to articulate the importance of important things
- the ability to convey your excitement for exciting things
You need all three to be much of a leader.
You can be a good leader and a bad manager. If you can’t be a good manager, you’ll have to be a good leader. If you can’t be a good leader, you’ll have to be an awfully good manager.
Leadership is important, because with leadership, you can get a bunch of people to work towards a common objective. Even if the common objective isn’t well-chosen, this situation will turn out better than everyone pursuing a different goal. So leadership ability is heavily rewarded in business.
To an extent, leadership is trainable. Courage may not be trainable, but you can at least learn to use all the courage you have. With practice, you can improve your ability to convey your excitement somewhat. With practice, you will get much better at explaining why something is important.
Leadership is not the same thing as charisma. Charisma is the knack for making people like and support you, in the absence of reasons for doing so. Leadership is the knack for making people like and support the plan, usually with the aid of enthusiastic yet careful explanation of the reasons for doing so. Many people think they can’t become good leaders because they aren’t charismatic. But charisma is over-rated. Many leaders have little charisma, and there is some evidence that charisma is actually detrimental to management ability. In any case, getting people to like you isn’t nearly as important as getting them to enthusiastically do what must be done. Charismatic leadership is indispensable to cults, but it’s optional otherwise, and often not the best option.
1.10.2006
Eleven Simple Steps to a Better Career Than You Deserve
Here are eleven simple rule for success in business. Well, they may be simple, but that doesn't make them easy. But it's nice to have a short list to work from, and most of the time business really is about this simple. Simple rules do oversimplify, of course, but they can be used in personal experiments. For example, pick a rule and emphasize it for a day or a week.
1. At the end of every day, ask yourself, "What did I learn today?" Be brutally honest. You might write down the answers, but if so never mention this to anyone.
2. Keep a list of commitments you’ve made and commitments made to you. It’s OK to do this in a very public way.
3. Keep your commitments. Every one. Don’t commit to what you can’t or won’t do.
4. When you are unsure if you should say something, say nothing.
5. When you are unsure if you should take offense at something, do not take offense.
6. Be loyal to your subordinates and your superiors. Only then will you see true loyalty in return. If you are hard to replace, your boss will probably keep you even if they do not like or trust you. However, if your subordinates do not like or trust you, eventually they will destroy you. The worst thing a boss can normally do is fire you. (Sometimes this is also the best thing a boss can do.) Your subordinates can destroy your career.
7. Consistently focus on the fundamentals: you will succeed even if you are not particularly talented. If you are particularly talented, you will still fail if you do not focus on the fundamentals.
8. Make everything as simple as possible. Start out with too simple and add complication ‘til it works.
9. Try to see the best in people. It’s natural to compare other people’s worst behavior to your best behavior, but you can’t afford that distorted view. When people do dumb or mean stuff, take the action you think needs to be taken, but have some mercy in your opinion of them.
10. Avoid speculating on effort estimates. Your speculations will be treated as commitments. Every time you break this rule, it will come back to haunt you. Every time.
11. If you have a sense of humor, tastefully show it.
1. At the end of every day, ask yourself, "What did I learn today?" Be brutally honest. You might write down the answers, but if so never mention this to anyone.
2. Keep a list of commitments you’ve made and commitments made to you. It’s OK to do this in a very public way.
3. Keep your commitments. Every one. Don’t commit to what you can’t or won’t do.
4. When you are unsure if you should say something, say nothing.
5. When you are unsure if you should take offense at something, do not take offense.
6. Be loyal to your subordinates and your superiors. Only then will you see true loyalty in return. If you are hard to replace, your boss will probably keep you even if they do not like or trust you. However, if your subordinates do not like or trust you, eventually they will destroy you. The worst thing a boss can normally do is fire you. (Sometimes this is also the best thing a boss can do.) Your subordinates can destroy your career.
7. Consistently focus on the fundamentals: you will succeed even if you are not particularly talented. If you are particularly talented, you will still fail if you do not focus on the fundamentals.
8. Make everything as simple as possible. Start out with too simple and add complication ‘til it works.
9. Try to see the best in people. It’s natural to compare other people’s worst behavior to your best behavior, but you can’t afford that distorted view. When people do dumb or mean stuff, take the action you think needs to be taken, but have some mercy in your opinion of them.
10. Avoid speculating on effort estimates. Your speculations will be treated as commitments. Every time you break this rule, it will come back to haunt you. Every time.
11. If you have a sense of humor, tastefully show it.
1.07.2006
Success
Success is a story. The story is always the same:
1. There is a problem.
2. You tackle this problem.
3. You do something clever, brave, or difficult.
4. The problem is solved.
You need to build up a collection of such stories. If they are good stories, others will tell them. The stories become part of the mythos of your company (or even your industry), and you acquire a positive reputation.
Try to get assigned work that is likely to result in a good success story, and avoid work that is likely to result in no story or a failure story. (There are failure stories too. They are same, except steps 3 and 4 are missing.)
There are things you can do that are beneficial to the company and enjoyable, but have no potential for a good success story. You can sometimes indulge in such work, but if you do too much of this kind of stuff, you’ll get a reputation as someone who isn’t going anywhere, who’s career has maxed out. Maybe it’s true. If you are happy doing your current job until you retire, go ahead and seek out “invisible” work.
What makes for a good story? Things that are related to the company’s core business makes for better stories. People understand the core business better and care about it more, so there is more dramatic potential. The problem you attack must be important. There must be a real threat to the business or a real opportunity to seize. Stories involving hated competitors or strongly desired yet previously unattainable customers are always good. The work must appear difficult or risky, so that the hero (you) appears brave. It’s best if the project is a sure bet and only appears risky, but it’s not likely that people will be fooled. You will probably have to show at least a little actual bravery. Really, courage in business is rare enough that courageous failure isn’t punished nearly as much as people think. Go ahead and show some spine. After all, it’s not like you will be literally executed if you fail.
1. There is a problem.
2. You tackle this problem.
3. You do something clever, brave, or difficult.
4. The problem is solved.
You need to build up a collection of such stories. If they are good stories, others will tell them. The stories become part of the mythos of your company (or even your industry), and you acquire a positive reputation.
Try to get assigned work that is likely to result in a good success story, and avoid work that is likely to result in no story or a failure story. (There are failure stories too. They are same, except steps 3 and 4 are missing.)
There are things you can do that are beneficial to the company and enjoyable, but have no potential for a good success story. You can sometimes indulge in such work, but if you do too much of this kind of stuff, you’ll get a reputation as someone who isn’t going anywhere, who’s career has maxed out. Maybe it’s true. If you are happy doing your current job until you retire, go ahead and seek out “invisible” work.
What makes for a good story? Things that are related to the company’s core business makes for better stories. People understand the core business better and care about it more, so there is more dramatic potential. The problem you attack must be important. There must be a real threat to the business or a real opportunity to seize. Stories involving hated competitors or strongly desired yet previously unattainable customers are always good. The work must appear difficult or risky, so that the hero (you) appears brave. It’s best if the project is a sure bet and only appears risky, but it’s not likely that people will be fooled. You will probably have to show at least a little actual bravery. Really, courage in business is rare enough that courageous failure isn’t punished nearly as much as people think. Go ahead and show some spine. After all, it’s not like you will be literally executed if you fail.
1.06.2006
Loyalty
Loyalty is always reciprocal. If you are loyal to your people, they will follow you through fire. Be loyal to your boss. If you don’t think your boss is loyal in return, get a new boss.
Loyalty means you stand by people, even at personal cost. It doesn’t mean you mindlessly back people up when they’re wrong. It means you give them honest answers, and keep your commitments. It means you take their objectives to be, in part, yours.
Loyalty plus success is how you build a faction that will support you. It’s how you make a name for yourself. If you have successes, but there is no mutual loyalty between you and your people, you will not make a name for yourself. This is because it is the words others say about you that makes your reputation. People will stick up for you if there is loyalty between them and you. This is a very important idea in building one’s career. A career is success stories, loyalty, and little else.
Loyalty means you stand by people, even at personal cost. It doesn’t mean you mindlessly back people up when they’re wrong. It means you give them honest answers, and keep your commitments. It means you take their objectives to be, in part, yours.
Loyalty plus success is how you build a faction that will support you. It’s how you make a name for yourself. If you have successes, but there is no mutual loyalty between you and your people, you will not make a name for yourself. This is because it is the words others say about you that makes your reputation. People will stick up for you if there is loyalty between them and you. This is a very important idea in building one’s career. A career is success stories, loyalty, and little else.
...a young prince must be prudent,
giving freely while his father lives
so that afterwards in age when fighting starts
steadfast companions will stand by him
and hold the line. Behavior that’s admired
is the path to power among people everywhere.
- Beowulf
1.01.2006
Listening
Listening is a big part of your job. In most conversations, you create more value by listening than by speaking. You must convey information and inspire, but you must also listen.
Here’s an interesting experiment: record a conversation of ten or fifteen minutes with someone, using one of those microrecorders. Ask the person’s permission first before you record. Normally, this works better if you don’t choose someone from work for the experiment. At the end of the recording, write down what percentage of the time you think you were speaking versus the other person, how many times you cut them off in mid-sentence, and how many times they cut you off. At your convenience, listen to the tape twice.
The first time, use a stop watch to record the fraction of the time you spoke. The easiest way to do this is to measure the total conversation length with your wristwatch, and use the stopwatch to record your speaking time. Wait for the other person to start speaking before you mark the end of your speaking time, so the entire conversation is spit into your speaking time and their speaking time.
The second time, count the number of times each of you interrupts the other.
If you are a man and want a humbling result, record a conversation with a woman.
Here’s an interesting experiment: record a conversation of ten or fifteen minutes with someone, using one of those microrecorders. Ask the person’s permission first before you record. Normally, this works better if you don’t choose someone from work for the experiment. At the end of the recording, write down what percentage of the time you think you were speaking versus the other person, how many times you cut them off in mid-sentence, and how many times they cut you off. At your convenience, listen to the tape twice.
The first time, use a stop watch to record the fraction of the time you spoke. The easiest way to do this is to measure the total conversation length with your wristwatch, and use the stopwatch to record your speaking time. Wait for the other person to start speaking before you mark the end of your speaking time, so the entire conversation is spit into your speaking time and their speaking time.
The second time, count the number of times each of you interrupts the other.
If you are a man and want a humbling result, record a conversation with a woman.
Subscribe to:
Posts (Atom)
