In High Output Management, Andrew Grove, former CEO of Intel, shares what he’s learned about management over his decades-long career. One of his main insights is that manufacturing principles, which focus on output and productivity, are very applicable to yet rarely used in management.
In this summary, you’ll learn:
The output of some processes, such as manufacturing, is very obvious—at a factory that makes breakfasts, the output is breakfast. Your own output as a manager, however, is less concrete. It’s not the output of your own individual work; it’s the output of all the teams you influence. Their output is determined by the activities you do and how much of an effect—or leverage—these activities have.
Written mathematically:
Your managerial output = team outputs = output of your team + output of teams you influence = (managerial activity #1 × leverage of activity #1) + (managerial activity #2 × leverage of activity #2) + …
Therefore, you can increase your managerial output by increasing the output of your team or the teams you influence, and/or increasing the value of your activities or your leverage.
To improve the output of your team, you need to choose the management style that will best foster each member’s high performance. The deciding factor regarding which style to use is task-relevant maturity (TRM)—how educated, experienced, and mentally ready an employee is to do a particular task in particular circumstances.
Here’s how you should manage employees with the following TRM levels:
Low: Provide structure and strong guidance on operational values. Give your team detailed directions about exactly how, when, and what needs to be done.
Medium: Provide some structure and guidance but expect the team to provide some of it themselves. Instead of managing prescriptively, exchange ideas and communicate.
High: Provide minimal structure—the team will provide the bulk of it themselves. Don’t involve yourself with the team’s work; just make sure objectives are clear. This style is advantageous because it allows the manager to delegate to their team members and be confident they will successfully complete the task.
Always monitor your team, no matter what the TRM level is. If the TRM changes, adjust your management style, up or down.
You can also increase your output by choosing to do valuable and/or high-leverage managerial activities:
1. Information-gathering involves collecting data about both the interior and exterior workings of your company (for instance, about customer complaints or team priorities). Do this by having conversations with people both inside and outside the company (competitors, customers, and so on), and by reading written information such as memos. Use both sources and check them against each other—verbal information is quicker to access but often incomplete, and written information is more thorough but slower to reach you.
2. Information dissemination involves supplying your organization and the people you influence with hard facts and also communicating goals, priorities, and preferred methods of doing things. Being informed allows people to do their jobs more efficiently.
3. Decision-making involves either making a decision yourself or helping others make decisions by sharing your opinion, fostering debate, and approving or vetoing ideas. Keep in mind that whoever’s highest in the organizational chart isn’t necessarily the person who has the most hands-on knowledge about a given issue, so ideally choose the decision-maker from the lowest competent level.
4. Nudging is a softer version of decision-making (and a stronger version of information dissemination) in which you influence people but don’t command them.
5. Role modeling is when you share your values and the kind of behavior you want to see in your subordinates by demonstrating them.
6. Training involves personally teaching your employees how to do the tasks they need to be able to perform. You should do this yourself rather than hiring external trainers because it’s an inherently high-leverage activity (you can affect many people at once), and because training works best when it’s tailored to company practices and the instructor is an authoritative role model.
7. Motivating involves creating an environment in which people’s need for status, or more ideally, self-actualization, is unmet (attempting to meet these kinds of needs provides longer-term motivation than meeting other kinds of needs, such as survival needs). If an employee is well-trained and capable of doing her job, the only reason she’s not performing is that she lacks motivation.
8. Giving performance reviews involves improving the performance of your subordinates (not just assessing it). To do this:
9. Hiring consists of interviewing a candidate and trying to determine from her self-assessment if she’ll be a good fit, as well as talking to references.
10. Retaining involves talking valued employees out of quitting. When a valuable employee decides to quit because she feels unappreciated, as a manager, you have failed. In this situation, immediately speak with the employee to show her she is appreciated. Then, brainstorm ways to keep her at the company, even if it involves transferring her to a different department and no longer working with her yourself.
11. Planning is coming up with actions that need to be done now to affect something in the future. To plan, evaluate your current status (what tasks you have in progress and what capabilities you have left over) and predict if maintaining this status will enable you to meet future demands on you and your time. If your status won’t meet the demand, adjust your planned workflow to make them match up.
Because an activity’s value is multiplied by leverage, if you increase the leverage of an activity, you can increase your output by a large factor.
There are three techniques to increase leverage:
1. Increase the leverage of existing activities. You can do this by affecting more people and their work, affecting people for a longer period, facilitating automation, or simplifying work (for example, removing unnecessary steps from workflows).
2. Stop doing activities with negative leverage. Don’t micromanage, waffle, or hold a bad attitude. All of these things will make it more difficult for your reports to get their work done and will decrease their output.
3. Adjust the number of reports you have. For maximum leverage as a manager, you should have 6-8 subordinates (or be part of 6-8 coordinating or planning groups) so you can give them each half a day of your time per week. (Half a day is ideal because it gives you enough time to effectively monitor but not enough time to micromanage.)
Like output, there’s a formula to describe managerial productivity (how efficient a manager is):
Managerial productivity = managerial output ÷ time = activity ÷ time required for the activity You can improve your productivity using the strategies to improve output (because activity and leverage are part of the equations) and one additional strategy: Speed up how long it takes you to do your activities.
To learn how to speed up, we’ll take inspiration from production manufacturing because it uses excellent time management techniques.
First, we’ll look at strategies directly inspired by the three steps of manufacturing. Then, we’ll look at some more general strategies used in manufacturing.
There are three steps to production in a factory:
1. Process, which is transforming raw materials into product parts. To determine the fastest, most cost-effective way to produce products in a factory, manufacturers first determine throughput times (how long it takes to prepare individual product parts) and then offset them (start them at staggered times so they’ll all be completed at the same time and ready for assembly). The schedule is arranged around the product part that takes the longest or is the most complicated to make.
Similarly, as a manager, arrange and offset your work schedule (and the schedule of your team) to ensure that all tasks involved in a particular project complete at the same time (in other words, the project deadline).
2. Assembly, which is attaching the product parts together (or, in a business context, bringing each part of a project together).
3. Testing, which examines the raw materials, parts, or the final product (or, in a business context, project) for flaws. Material gains value as it passes through each production stage (for example a whole breakfast is more valuable than a raw egg). Therefore, you should always try to catch problems at the lowest value stage to save yourself money (scrapping a faulty part is less expensive than scrapping a faulty product). To do this, test at three different points during production:
In general, the best way to test is “monitoring,” which involves taking samples and testing them while the rest of the material moves on to the next stage. If the sample testing turns up a problem, you can stop the rest of the process.
There are six manufacturing strategies that you can apply to management to improve your efficiency:
1. Use indicators, which are measurements that tell manufacturers (or managers) about what’s going on in the production process (or administrative process) and predict or provide information about output. To choose which indicators to pay attention to, ask yourself what information you want to know first-thing every day to head off potential problems.
2. Forecast output. Most manufacturers predict how many orders they’re going to get and build enough inventory to satisfy their predictions (as opposed to building only after receiving an order, which results in long wait times for a customer). As a manager, you can forecast too—predict demands on your time and plan accordingly.
3. Use proven workflows. Manufacturers don’t waste time coming up with a new way to do something if there’s already a good existing way. Don’t reinvent the wheel as a manager either.
4. Batch. All processes include set-up time, and if you can do all the activities that require the same set-up together, you only have to spend the set-up time once.
5. Don’t overload capacity. Overloads might cause bottlenecks and materials might have to be aborted at a higher-value stage. To avoid overloads as a manager, refuse projects you don’t have time for.
6. Distribute workload. Factories try to evenly distribute their workload over time, which managers can benefit from too. The main threat to an even managerial workload is interruptions because they unexpectedly increase workload. To manage interruptions:
Meetings, like manufacturing strategies, can be an effective time-management tool. Meetings are a medium in which you can do managerial activities, many of which need to be carried out in person.
There are two types of meetings, and if you use your meeting time perfectly, you’ll only ever need to hold the first kind:
1. Process-oriented. The goal is to exchange expertise and information. These meetings take place according to a regular schedule. There are three subcategories:
These meetings save time because they allow you to pre-empt problems. If an attendee sees a potential problem and alerts you right away, you have time to address it before it becomes a larger issue.
2. Mission-oriented. The goal of these meetings is to solve a problem by making a decision. In theory, these meetings should never be called because all problems are preempted in the process-oriented meetings. In practice, however, even if things operate well, about 80% of problems will be handled in process-oriented meetings and the remaining in mission-oriented meetings.
Managing a whole organization, as opposed to a team, involves two elements:
Organizational structures are arrangements of the organizational chart: for instance, they show which unit in the business does what, and how these units work together (if they do at all).
There are three organizational structures:
1. Functional. In this centralized structure, individual business units are only responsible for their unit-specific tasks. Any function that they share with another unit (for example, human resources), is handled by a functional group that handles the shared function for all units. The advantages of this structure are economies of scale, the opportunity to share expertise (experts can share their knowledge across the whole company, not just their unit), and units having the ability to focus on their work, not administration. The disadvantages are increased bureaucracy when trying to get help from functional groups, and resource competition between the distinct units.
2. Mission-oriented. In this decentralized structure, every business unit is responsible for both its unit-specific tasks and all the other tasks that come with running a business, such as hiring, purchasing, offices, and so on. Each unit reports to a regional office or the corporate executive office. The only advantage of this structure is speed of responsiveness; units don’t have to wait on other departments to do anything. The disadvantages are the inability to collaborate with other units and redundancy (for example, each unit has an HR department when really, the company only needs one).
3. Hybrid. This structure is a combination of functional and mission-oriented. Individuals can report to multiple supervisors and be part of multiple organizational charts, which allows them to fulfill both functional and mission-oriented responsibilities.
Ideally, the hybrid structure harnesses all the advantages of each system and dispenses with the disadvantages. According to Grove’s law, as businesses grow, they’ll eventually need to transition to this structure because as they get bigger, they have more things to keep track of and need the advantages of a functional structure as well as those of the mission-oriented approach.
There’s no universally optimal way to control people’s behavior—the most effective approach always depends on the circumstances.
There are three different control modes applicable to the workplace:
1. Free-market forces. In this mode, people’s behavior is controlled by price (whoever is buying wants to pay the lowest price, and whoever is selling wants to sell at the highest price possible) and self-interest. No management is needed because everyone openly does whatever is in their best interest.
2. Contracts. In this mode, people’s behavior is controlled by mutually agreed-upon guidelines that outline what each party will do to what standards and who has the right to monitor work. Contracts are useful when exchanging goods or services that don’t have a defined value, like the contribution of an individual engineer. Management is needed in this method—management helps set and enforce the contract guidelines.
3. Culture. In this last mode, people’s behavior is controlled by culture—people believe and trust that the whole group shares ways of doing things, values, and goals. Culture is a useful control method when it would be impossible to define dollar values of or contracts surrounding behavior. Management’s role is to develop and establish culture by explaining ways of doing things, values, and goals and also visibly demonstrating them.
In High Output Management, Andrew Grove, former CEO of Intel and 1997 Time Man of the Year, shares what he’s learned about management over his decades-long career. One of his main insights is that manufacturing principles—which have been honed over hundreds of years and focus on output and productivity—are very applicable to yet rarely used in management.
In this summary, you’ll learn:
The book is relevant to all managers but may be particularly useful to middle managers because there are fewer resources available for them (there are many courses for front-line managers, and most business schools focus on training CEOs). Middle managers are all managers in between front-line managers and the people at the top of the organizational chart. The term also includes “know-how” managers who don’t formally have reports but influence the work of others by sharing their knowledge and skills.
(Shortform note: We’ve reorganized the book’s material for concision and clarity.)
The original edition of High Output Management was published in 1983. Grove wrote a new introduction to the book in 1995 because two major events took place that changed the managing game:
Event #1: The world became more globalized. Businesses now had competitors all over the world, not just locally, and if they didn’t adapt, they died. Because so many companies could supply the same products, speed became a major competitive advantage.
Competition for jobs also increased—managers have to work harder to maintain their personal competitive advantage. Grove recommends asking yourself the following questions to reflect on your work. If you answer no to any of them, your career as a manager might be in danger:
Event #2: Email was invented. It increased the speed and ease of communication and made the managerial activity of disseminating information less critical to the job. Keep this in mind when reading the sections on managerial activities and meetings.
The output of some processes, such as manufacturing, is very obvious—at a factory that makes breakfasts, the output is breakfast. Your own output as a manager, however, is less concrete. It’s not the output of your own individual work; it’s the output of all the teams you influence. Their output is determined by the activities you do and how much of an effect—or leverage—these activities have.
Written mathematically:
You managerial output = team outputs = output of your team + output of teams you influence = (managerial activity #1 × leverage of activity #1) + (managerial activity #2 × leverage of activity #2) + …
Therefore, you can increase your managerial output by increasing the output of your team or the teams you influence, and/or increasing the value of your activities or your leverage.
First, we’ll look at how to increase the output of teams by choosing an appropriate managing style. Then, we’ll look at high-leverage managerial activities that you can engage in to increase your output. Finally, we’ll look at how to increase leverage.
To improve the output of your team, choose the management style that will best foster their high performance. The deciding factor in this decision is task-relevant maturity (TRM)—how educated, experienced, and mentally ready an employee is to do a particular task in particular circumstances.
Here’s how you should manage employees with the following TRM levels:
Low: Provide structure and strong guidance on operational values. Give your team detailed directions about exactly what needs to be done, how, and when.
Medium: Provide some structure and guidance but expect the team to provide some of it themselves. Instead of managing prescriptively, exchange ideas and communicate. (Note: This mode is the most difficult for most managers because hitting a balance is more challenging than meeting an extreme.)
High: Provide minimal structure—the team will provide the bulk of it themselves. Don’t involve yourself with the team’s work; just make sure objectives are clear. High TRM is advantageous because:
Always monitor your team, no matter what the TRM level is. If the TRM changes, adjust your management style, up or down.
None of these management styles are better or worse than others. Many managers feel the prescriptiveness of low-TRM management isn’t nice or sophisticated, but your goal isn’t to be these things—it’s to be effective.
Whether you foster friendships with subordinates is up to you. It’s easier to care about and communicate with someone you’re friends with but harder to revert to a formal relationship if necessary because ordering a friend around can be awkward. Additionally, close personal relationships don’t necessarily result in communicating management relationships.
Grove has seen friendship with subordinates work and he’s also seen discipline end friendships. To figure out if you can handle it, imagine that you’re giving a critical performance review to an existing friend. If the thought makes you uncomfortable, don’t become friends with your subordinates.
You can also increase your output by choosing to do valuable and/or high-leverage managerial activities. We’ll cover five activities in this chapter and the rest in subsequent chapters.
Information-gathering involves collecting data about both the interior and exterior workings of your company. You’ll use what you learn from this activity to inform all of your other managerial activities.
There are two main sources of information, and you need to gather both:
1. Verbal, via conversations with people both inside and outside the company (competitors, customers, and so on). Verbal information is more up-to-date than written information and therefore valuable, but often insufficient or even incorrect because in conversation, people don’t always mention all relevant context.
There are two good ways to get access to verbal information:
2. Written, via memos, reports, and so on. Written information travels more slowly than verbal information but is necessary and valuable because:
You should have multiple sources of both verbal and written information so that you can compare the different accounts to each other to confirm they’re accurate.
A second managerial activity is supplying information to your organization and to the people you influence. Information includes hard facts but also goals, priorities, and preferred methods of doing things. If you don’t share this information, then your reports won’t know how you want things done and their output will be sub-par.
Sharing information also contributes to the development of culture, which we’ll look at more in later chapters.
The third managerial activity is decision-making. As a manager, you have two decision-making roles:
To know which to do, consider the two kinds of power:
In traditional industries with a strict organizational hierarchy, decision-making responsibilities were assigned based on position power, so the higher-ups in the chart would make the decisions. However, in knowledge industries, a person with position power may not have as much knowledge power as someone lower in the chart because they’re more distanced from day-to-day operations, and the faster-moving the industry or customer preferences, the larger the disconnect.
Therefore, the best way to make decisions is to harness both kinds of power and make the decision at the lowest competent level—lowest because lower people are closer to the action and technical know-how, and competent because technical knowledge doesn’t always come with judgment or experience. As a middle manager, you’re uniquely positioned to do this because you’re in contact with both power-holders.
If you can’t find people who have both the technical knowledge and judgment to make, or help you make, the decision, then create a decision-making group that includes a mix of people: some of whom have technical expertise, and some of whom have proven strong judgment. It’s fine to include someone senior, but she must act as an equal to everyone else involved so that everyone feels comfortable participating in the discussion rather than deferring to the senior leader.
It’s important to adhere to a standard decision-making process for two reasons:
There are four steps to Grove’s ideal decision-making model:
Step #1: Address the Six Pre-Decision Questions
Before starting the decision-making process, to promote an optimal, efficient decision, you must give everyone involved in the process the answers to the following six questions:
1. What is the decision?
2. What is the deadline?
3. Who will make the decision?
4. Whom do you need to talk to before making the decision?
5. Who will approve or reject the decision?
6. Who needs to know the decision?
Step #2: Encourage Free Discussion
Discussion should cover all points of view and all relevant information because the best decision will come from considering the opinions of all knowledgeable people. Everyone should get to speak freely. Encourage disagreement—it’s common for participants to stay out of the discussion until one view seems likely to win and then jump in to support this view, but this behavior won’t produce the best decision because it results in people withholding their true opinions.
Free discussion ends when you’ve heard and considered all the information. At this time, push for a consensus, or make the decision yourself (you’ll be well-informed thanks to the free discussion). This timing is important:
For example, the Intel groups considered all the information about the new plant’s possible locations, including costs, geography, and traffic patterns. They thought it would work best to build the new plant beside the old plant but keep the new plant small to avoid huge costs.
Step #3: Explain the Decision Clearly
Once something is decided, it should be explained as clearly as possible to everyone who will be affected by it.
People tend to be vague about controversial decisions to avoid arguments with those who disagree with the choice, but you should do the opposite. Being vague doesn’t prevent an argument, it just postpones it, and people don’t like being kept in the dark.
If the decision is very different from what was on the table (for example, if Grove had canceled the construction of the plant), announce it, give people some time to digest it, and then later, talk about it with everyone. Some time for reflection and the opportunity to share feedback will help people accept the decision.
Step #4: Demand Full Support
Everyone involved in the decision must support it. They don’t have to agree with it—it’s not always possible to get agreement on everything because everyone has differences of opinion—but they must commit to executing it.
This four-step process comes with challenges:
1. Unnaturalness. Middle managers often find this model awkward because they struggle to express their views, make hard decisions, and get behind decisions they don’t agree with.
2. Emotions. It’s hard for people to separate their emotions from decision-making, and people tend to get even more emotional when they have to make decisions with people they don’t know. If there’s a big divide between knowledge and power, everyone can feel uncertain. People might fear sounding stupid and not speak, or fear being overruled and therefore feel embarrassed.
To manage your emotions, remember that you, like all people, are smart and have willpower.
3. Peer-group syndrome. When in a group of peers, most people are reluctant to express an opinion because they don’t want to say something that the rest of the group doesn’t agree with. Therefore, people talk around in circles until they feel confident that the group is leaning towards something (or if this doesn’t happen, they talk forever). One member will state the perceived consensus weakly using first-person plural, so it’s not attached to them personally. If they’ve read the room right, everyone else will start supporting the position too. This is disadvantageous because it cuts the discussion short, and without examining all points of view, it’s harder to make a good decision.
There are two ways to avoid this circling:
The fourth managerial activity is nudging, which is a softer version of decision-making (and a stronger version of disseminating information) in which you influence people but don’t command them.
Nudging and decision-making should be distinct. You’ll probably nudge a dozen times as much as you decide.
The fifth activity is role modeling, in which you share your values and the kind of behavior you want to see by demonstrating it. Telling people what you want in a memo will never be as effective as showing them visibly. You can do this in whatever way fits your leadership style—such as working one-on-one or giving a presentation.
In the last chapter, you learned about five of the 11 managerial activities. In this chapter, you’ll learn about another two: training, and creating a motivational atmosphere.
There are only two reasons an employee doesn’t do her job: Either she’s untrained and therefore incapable, or she’s not motivated. To figure out which it is, ask yourself if she could do the job if her life depended on it. If she couldn’t, the problem is capability; if she could, the problem is motivation.
Employees who aren’t trained properly, no matter how good their intentions, can cause all sorts of problems for a company ranging from bad performance to excess costs to life-threatening situations.
You, as a manager, should train your subordinates (and possibly their subordinates) yourself rather than hiring outside training specialists because training works best when:
To figure out what to train your people in:
To develop a training program, start small, with a course that includes fewer than five lectures, because teaching will likely be more difficult than you think. When preparing the course:
At Intel, employees spend 2-4% of their time in classroom learning and most of the teaching is done by the company’s managers. Managers teach over 50 classes on a variety of subjects.
There are two types of training at Intel:
The more people who require training, the longer it takes and the more expensive it is, but it’s worth it.
If an employee’s poor performance isn’t due to a lack of training, it may be due to a lack of motivation. Motivation comes from the inside, so you can’t instill it in or force it on someone. What you can do is foster an environment where people can become motivated and perform well.
According to Abraham Maslow, people are motivated by the desire to satisfy their needs. Once a need is met, motivation ceases. Therefore, to create motivation, you need to create a climate in which some of people’s needs are constantly unsatisfied.
People have many needs and they fall into a hierarchy—only once you’ve satisfied your most important and basest needs do you start trying to satisfy your higher needs.
Here are the needs, in order of lowest to highest:
The first three needs inspire employees to get a job and report for work. All three of them are possible to satisfy, so when they’re met, motivation ceases.
The last two needs encourage high performance. Status is satisfiable, so motivation will eventually cease once an employee reaches a certain level of success, but self-actualization is limitless because there’s no concrete goal attached to it. Therefore, as a manager, you need to get everyone to the self-actualization stage to tap into their best performance.
To help everyone reach self-actualization:
1. Assess where people fall in Maslow’s hierarchy of needs. One way to do this to look at how they view money and raises. If a subordinate gets a raise and she’s concerned with the absolute value, then she’s working at the survival or security level. If she’s concerned with the relative value (how her raise compares to her coworkers’), then she’s likely after status or self-actualization and is using money as a measure of achievement.
2. Reassess hierarchy position from time to time. People can move both up and down Maslow’s hierarchy.
3. Base achievement recognition on output. Only attach status and self-actualization to achievements that are output-relevant. Otherwise, people will be motivated to work towards things that don’t benefit the company, such as competing for the nicest office chair.
4. Consider the effects of fear. In the past, fear of punishment was the main motivational strategy. (If people didn’t work, they didn’t have the money to eat. Their only option would be to steal food, the punishment for which was hanging. Therefore, people were motivated to work to avoid the punishment of death.) At the time of writing, fear was being replaced by more humane motivational strategies. However, fear is still present in status and self-actualization, though it’s usually internally-generated fear of failure. Fear can be motivating—it encourages people to perform—but it can also become an obsession, which results in a reluctance to take risks because they come with the danger of failure.
5. Create opportunities for people to test themselves. People will produce greater output trying to achieve things that are just beyond their reach (self-actualization is always a little out of reach).
6. Make work mimic competitive sports. Competition is highly motivating, so if you can tap into this, you can increase performance. To do this:
7. Provide feedback. When striving to meet the need for self-actualization, motivation is generated by the drive to improve, and it’s impossible to improve without feedback. People provide their own internal feedback—for example, when a professional violinist is practicing and hears herself play a wrong note, she can go over the section again until she gets it right—but also require external feedback. We’ll look at how to provide this in the next chapter.
You’ve now learned about seven of the 11 important managerial activities. In this chapter, we’ll look at the eighth: giving performance reviews.
The primary goal of a performance review is to improve performance (not just assess it). Giving performance reviews is a high-leverage activity because it affects the recipient’s performance for a long time—she often recalls the feedback. Performance reviews should be used in all organizations regardless of size because performance matters to every organization.
Managers don’t naturally write good reviews for three reasons:
First, we’ll look at a four-step process for how to handle the first two difficulties and write good performance reviews. Then, we’ll look at how to handle the third difficulty.
Before you can assess performance, you need to determine your expectations so that the review can focus on whether or not your subordinates met them. (Failing to do this is the major problem with most reviews.)
As you come up with your expectations, remember that output is an important marker of performance. Consider the recipient’s actions, not just her results, because:
1. Often, there’s a delay between doing an activity and seeing the output. The output might not accurately reflect the results of the recipient’s activities during the period the review covers.
For example, when Grove was reviewing one of the managers who reported to him, the manager’s output measures were all strong—for example, sales had increased. There were some signs that not all was well—for example, the manager’s organization had higher-than-usual turnover—but Grove prioritized the output measures and gave the manager a good review.
The next year, the output measures had all declined. Grove looked carefully at the situation and realized that the offset between the manager’s activity and output was about a year. The output measures were only good the first year because they had been set up by previous employees and only come to fruition years later. Therefore, the manager hadn’t deserved the good review because the outputs didn’t reflect his actual performance. Interestingly, the manager’s performance had actually improved since the first year, even though the output measure didn’t reflect this at all: His work just hadn’t come to fruition yet.
2. You need to assess both short- and long-term projects. Long-term projects may not show output for years, so to evaluate them, consider the future gains and what they’re currently worth.
Once you’ve established your expectations:
1. Don’t ask the review’s recipient to write a self-review for you to reference because:
2. Gather data on the recipient’s performance by collecting meeting notes, progress reports, and so on.
3. On a blank sheet of paper, write down every thought you have about the recipient’s performance. Don’t edit them or order them by importance; just write down your stream of consciousness.
4. Once you’ve finished, look for broad themes. Probably, you’ll see evidence that the person does the same actions in various contexts.
Don’t focus on potential or appearances because it doesn’t matter how professional someone acts or how good she could be—for example, if the recipient is a manager and her team doesn’t perform, her performance is flawed.
5. Call these themes “messages” and make a list of them. Then, support the messages with examples from your worksheet. (Ideally, none of your messages are surprising because you monitor the recipient and provide guidance regularly. If you do discover something you weren’t expecting that might appear to the person as coming out of left field, include it anyway: They need to know about it to improve their performance.)
6. If the list of messages is too long for the recipient to remember, cut the less critical ones. You can probably mention them in a future review.
As discussed, the goal of the performance review is to improve performance. Therefore, after assessing performance, you need to provide task-relevant feedback—feedback that is directly related to work—to guide recipients toward improvements.
There are three formats for providing feedback in performance reviews:
In this type of review, which is very common, managers list negative feedback and then temper it with positive feedback. These reviews aren’t very effective at improving performance because:
In this form of review, you discuss and try to resolve a single fire-worthy performance problem.
Likely, when you bring up a major problem, the recipient will go through the following stages:
1. Refuse to acknowledge the problem exists. To show the recipient the problem is real, use examples and evidence.
2. Deny that it’s a problem. Again, use evidence to show that the problem is real.
3. Accuse others of causing the problem. In this stage, the recipient accepts that the problem is real but doesn’t think it’s her fault and therefore doesn’t plan to do anything about it. This tends to be the hardest step to get through. If she can’t get through it on her own, use your position power. Tell her that you understand she doesn’t see the problem the same way you do and that either of you could be right, but it’s your responsibility to the organization to give instructions.
4. Assume ownership of the problem.
5. Find a fix. You should help the recipient do this. It’s typically easier than the other steps because finding fixes is a matter of logic, and moving through the other steps is emotional, which is more difficult for some people to handle.
The recipient may not go through all the stages, and if she doesn’t move on from one, you can’t either (it’s impossible to talk about fixing a problem that someone refuses to admit even exists).
There are three ways this review can end:
The first outcome is ideal because if the recipient feels the same way about the problem as you do, she’ll likely be more motivated to solve it. However, the second option is acceptable too because while it might not be as emotionally comfortable to disagree, the problem will still get solved. The third option is not acceptable.
In this review, you give feedback to a high achiever. Because the goal of a review is to improve performance, always include suggestions for improvement, even if someone is already doing well. (When it comes to high performers, there’s a tendency to simply summarize the subordinate’s work instead of providing suggestions to improve.)
High performers do a lot for the organization, so if you can improve their performance even more, you’ll improve the company’s output significantly.
Once you've decided which type of feedback to provide, it’s time to deliver the review. You should give a recipient a written review in advance of your in-person discussion because:
Reason #1: It will give her time and privacy in which to process the information. She’ll be more prepared to discuss her review when the time comes.
Reason #2: If you give her the written review after the discussion, she might find something that surprises her because she wasn’t listening carefully in the meeting and get upset about it.
Reason #3: If you give the written review during the meeting, she may pay more attention to it than what you’re saying. Additionally, she won’t have time to prepare what she wants to say in response.
There are three important to-dos when it comes to delivering the in-person review:
1. Be honest and straightforward. The process won’t be helpful if you give incomplete or wishy-washy feedback. Handle both negative and positive feedback with the same honesty and straightforwardness.
2. Confirm that the recipient is absorbing your feedback. Just because you’re talking to someone doesn’t mean she’s listening or processing—if she’s emotional, thinking of what she’ll say in response, or is tuned out, she may not hear what you’re saying.
Use your senses to make sure you’re getting through to the recipient. For example, watch her as you’re talking to her. If her body language indicates she’s not listening, keep communicating. It’s your responsibility, not hers, to get the message across.
3. Focus on the recipient. The review is about her, not you. You’ll likely have strong emotions because giving a review can be nerve-wracking, but you need to control them.
Promotions should be based on performance because when you promote someone, you create a role model and show what qualities (high performance) you value.
Note that when you promote someone who’s exceeding expectations, her performance will actually decrease in the new position to merely meeting expectations. This is because she’s still learning the role and navigating new responsibilities. As she gets more experience, training, and motivation, her performance will improve and she’ll start exceeding expectations again.
You may fear promoting exceptional people because you don't want their performance to decrease, even temporarily. However, if you don’t promote them, you’re wasting a resource. Eventually, they lose motivation, stop exceeding expectations and drop back to only meeting them, and settle for being average.
If you promote someone to a position that she’s not even able to meet expectations in, it’s not the failure of the employee, it’s the failure of management—you made a bad judgment call about her abilities. In this situation, you can fire her (which Grove thinks is “dead wrong”), or, you can “recycle” her—return her to the position she held (in which she exceeded expectations) before she was promoted. She’ll probably be embarrassed, but if you’re straightforward and supportive she’ll likely get over it quickly. You might promote her again later, once she’s gotten her confidence back and you’ve better assessed her readiness.
Compensation, like promotions, should be based on performance to reinforce the importance of good performance.
There are three ways to determine base salaries:
No matter which system you use, salaries should max out because all jobs have an upper limit to their value. Once someone reaches the maximum, don’t keep giving raises on a schedule.
When it comes to paying managers you supervise, whose output is more difficult to determine because it's based on other people’s, you should create a performance bonus so that at least part of the managers’ compensation is based on team performance. This reinforces that a manager’s output is related to her team’s. The higher the total compensation, the higher the percentage of the bonus should be.
To build an effective performance bonus system:
For example, you might end up with a system where different factors contribute a percentage to the overall performance bonus. 5% could be determined by individual performance, 5% could be from the immediate team’s performance, and 5% from the company’s performance. This allows you to make each type of performance visible.
You can ask your subordinates to review you as a manager, but make it clear that their review of you is informal and advisory, while your review of them is your job. Never pretend you’re on the same level as your subordinates in a review.
There’s a step by step process to assessing performance.
Choose an employee to evaluate. Where could you find data on her performance? Consider looking at progress reports, meeting notes, and so on.
Write down every thought you have about your employee’s performance.
What broad themes do you see on the list? (For example, the employee might consistently miss details: Her emails are full of typos, she misses items in lists, and she skips agenda items when she runs meetings.)
If you give the employee too much feedback at once, she’ll probably be overwhelmed and won’t remember any of it. What are the most important three themes from your list above? Why?
In this chapter, we’ll look at the next two managerial activities: hiring new employees and retaining existing employees who have decided to quit.
First, acknowledge that hiring is hard. In activity #8, you learned how hard it is to assess the performance of people who already work closely with you even when you have output measures, specific examples of their work, and a confirmed environment. Evaluating the potential performance of a stranger is even more difficult. There’s no foolproof way to hire—even if you follow all of Grove’s recommendations below, you might still end up with a bad hire.
When hiring for a position, your goals should be to:
You have two resources with which to do this:
The goal of interviewing is to predict how an applicant would do at your company based on her self-assessment of her own abilities. There’s no way to avoid using self-assessment at the interview stage because you don’t have access to objective measures of skills, performance, and other factors. However, even though self-assessments can be inaccurate, they can still provide good information if, as the interviewer, you’re blunt and direct, because this usually inspires similar directness in the applicant.
Here are some tips for interviewing:
Tip #1: Aim to learn the following types of information:
Tip #2: Listen. The applicant should talk for 80% of the time so you can learn as much about them as possible.
Tip #3: If the applicant gets off topic, interrupt her and tell her you’d like to change the subject. You only have an hour or two—use it wisely.
Tip #4: Use the same language and make sure you both understand the meaning of any jargon you use.
Tip #5: Ask the applicant to consider a hypothetical situation. For example, when Grove was interviewing an accountant, he asked him to determine the cost of a finished wafer. Grove explained the wafer manufacturing process and answered questions about the process, since the accountant didn’t know how it worked, and the accountant worked out how to determine the price.
Tip #6: Ask the applicant if she has questions. This will show you if she prepared for the interview, if she researched the company, and what she’s looking for in a company.
Tip #7: Don’t use tricks—for example, giving applicants a broken chair and then seeing how they react when they fall. Instead, give a realistic first impression of yourself and the company because it takes first impressions a long time to fade, and encountering tricks during an interview suggests there will be tricks and traps on the job too.
When checking references, you want to get the same four pieces of information (see Tip #1) from references that you aimed to get in the interview. This will be difficult for two reasons:
To mitigate these challenges, try to develop a bond with the reference to get her to open up to you. Do this by finding something you have in common or stretching out the conversation to give you time to connect.
After talking to references, ideally, interview the applicant again to address anything that came up.
The last managerial activity is talking people out of quitting. When a valuable employee decides to quit because she feels unappreciated, as a manager, you have failed. Your failure doesn’t necessarily end with the departure of that one employee. The actions of good employees affect your other employees as well—people who respect and identify with the departing employee will be affected by what happens to her and potentially become demotivated and consider leaving themselves.
Usually, someone will tell you that she wants to quit when you’re in the middle of something important. You should:
1. Immediately talk to her. This probably won’t be your first reaction—the conversation will be uncomfortable and your instinct will likely be to wrap up whatever important thing you were in the middle of first. However, if your employee is quitting because she thinks you don’t care about her, pushing off the conversation until later only reinforces her belief.
2. Ask her why she’s decided to quit and listen. She’s probably rehearsed this conversation, so let her get through her script without interrupting, and only ask questions at the end. Her script probably didn’t touch on her real reasons for quitting, so try to get her to reveal those by encouraging her to talk more.
3. Don’t argue or try to change her mind yet. At this early stage, all you should do is ask for some time to come up with a solution.
4. Go talk to your supervisor—this is a major problem you need help with. Your supervisor will also probably be in the middle of something important, but you need to make her help you. While your subordinate quitting is more destabilizing for you, her departure will affect the whole company.
5. Brainstorm ways to keep the employee at the company. This can include transferring her to a different department—while you’ll lose her, the company won’t. As a manager, it’s your (and all managers’) responsibility to keep valuable employees with the company, even if you don’t get to work with them yourself.
6. Speak with the employee again and present your solution. She might counter by:
In this chapter, we’ll first look at the 11th and final managerial activity, planning—coming up with actions that need to be done now to affect something in the future. Then, we’ll look at delegating, which will help you make time to complete the 11 activities we’ve explored.
There are four steps to effective planning:
In management, demand comes from your environment, which is the other groups in your organization that closely interact with you.
To determine your demand, consider:
Then, consider the expected state of all these factors in the future to determine the difference between what people expect of you now and what they might expect in the future. This is called a difference analysis.
Don’t try to resolve anything yet—that will come in the next steps.
There are two things to consider when evaluating your current status:
Use the same vocabulary to describe status and demand. For example, if your demand is for “completed product designs,” refer to anything in-progress is a “partly completed product design.” That way, you can more easily compare your status and demand.
In this step, you’ll change what you’re currently doing so that you can meet future environmental demand. To do this, ask yourself two questions separately:
Decide on your strategy (a general summary of what you plan to do) and tactics (what you’ll actually do to make your strategy come true), keeping in mind how your actions will reconcile and how long this reconciliation will take.
Make sure to involve the people who will be responsible for implementing the plan. For example, Cindy made a list and a schedule of what steps the development engineers needed to do to produce enough documentation for her needs. She spoke with the development group’s manager to negotiate what was possible.
You only need to implement actions that need to be done in between the present moment and your next re-evaluation of the strategy you determined in step #3. For example, every year, Intel plans for the next five years. Because they plan annually, they only need to focus on implementing one year’s strategy, because they’ll look at the other four years and one new year in the next plan.
Don’t plan too often, or you won’t have time to get the feedback you need to adjust your plan before you have to implement it.
When planning for short-term concerns, you can use the “management by objectives (MBO)” technique. Because the timeline is short, you should already have a good idea of present demands on your time, so MBO focuses on steps 2 and 3 of the previous section. When using MBO, you need to determine only two things:
1. Objectives. Only choose a small number of objectives for yourself and your subordinates—if you have too many, your focus will be diluted, and none of them will be done well. Note that if your subordinates meet their objectives, this will result in your objective being met because their objectives are to execute tactics, which will lead to carrying out the strategy, which is your objective.
2. Milestones (indicators of pace). These need to include specific wording and dates so that it’s very clear whether they’re reached. The deadlines should be short so you can get feedback on your progress quickly.
Subordinates can still miss objectives even if they reach all the milestones on time. However, this doesn’t necessarily mean they’ve performed badly—they may have just focused on other, equally productive opportunities.
To give yourself time to do the valuable and high-leverage actions in this chapter and the previous ones, you’re going to have to stop doing low-leverage activities. You can do this by delegating—many lower-leverage activities do need to be done, but not necessarily by you. There are some criteria for delegating:
1. You and the person you’re delegating to need to know the same information and be in agreement about how to do the task. If the person you’re delegating to can’t do the task without you giving them step-by-step instructions, you’ll need to spend a lot of time giving directions, and the leverage of this time will be low.
2. You can keep the tasks you like doing as long as you’re consciously aware of what’s motivating your decisions to delegate or not to delegate. If you give away a task but can’t let go, you’ll spend time on it regardless, and your leverage will remain low.
3. Monitor the person you’ve delegated to. Even if you’re not doing the task yourself, you’re responsible for it getting done, so you can’t completely step back. To monitor:
(For more on monitoring, see the “Testing” section in Chapter 6.)
4. When possible, delegate the tasks you’re familiar with rather than the ones you aren’t because it will be easier to monitor something you already know how to do.
5. If you’re delegating decision-making, monitor the decision-making process. Ask the person to reflect before presenting a decision. Then, ask a question to monitor how well she reflected.
In the previous chapters, we looked at the activity term of the managerial output equation. In this chapter, we’ll look at the leverage term. Because an activity value is multiplied by leverage, if you increase the leverage of an activity, you can increase your output by a large factor.
There are three techniques to increase output when it comes to leverage:
For maximum leverage as a supervisory manager, you should have 6-8 subordinates so you can give them each half a day of your time per week. (Half a day is ideal because it gives you enough time to effectively monitor but not enough time to micromanage.)
Though know-how managers don’t have official subordinates, they should also aim to influence 6-8 people or 6-8 coordinating or planning groups.
If your organization’s hierarchy doesn’t provide you with the right number of subordinates, you can “act” as a lower role to get the right amount.
We’ve already looked at how to choose effective managerial activities. Now, we’ll look at how to make these activities more powerful, which you can do in five ways:
1. Affect a larger number of people. Often, you can do this by doing activities at the opportune moment—when you’re in the presence of or can influence many people at once.
2. Affect people for a longer period, ideally by doing something that only takes you a short amount of time.
3. Automate. Look for ways to help automate your subordinates’ work—machines can help people increase their output.
4. Simplify workflows. Often, workflows contain unnecessary steps that can be eliminated to streamline to process. To simplify:
Some managerial activities have negative leverage—they reduce the effectiveness of people on your team.
Three activities usually come with negative leverage:
1. Managerial meddling, which is when a manager takes over her subordinate’s work. It’s negative leverage because after it happens enough times, subordinates no longer take the initiative and instead of doing their own work let the manager do it instead. Meddling usually comes from a manager having too much free time. When you have discretionary time, start a backburner project instead of interfering in others’ projects.
2. Waffling, which is when a manager procrastinates making a decision that will affect others. Her subordinates can’t do anything until the decision is made, so their effectiveness is hampered.
3. Holding a bad attitude. When a manager is depressed or negative, the feeling spreads to her subordinates and discourages them from giving their all.
In the previous part, we learned an equation to describe output and how to influence the various factors in that equation to increase output. In this part, we’ll look at managerial productivity and methods to improve it, thus increasing your effectiveness as a manager even further.
Like output, there’s a formula for managerial productivity:
Managerial productivity = managerial output ÷ time = activity ÷ time required for the activity
You can improve your productivity using the strategies to improve output (because activity and leverage are part of both equations) and one additional strategy: Speed up how long it takes you to do your activities. Your own time is the only limited resource—unlike manpower or capital equipment, you can’t get more of it no matter how much you might be willing to pay. As a result, Grove thinks that time management is the most important part of leadership.
To learn how to speed up productivity, we’ll take inspiration from production manufacturing because it uses excellent time management techniques and they’re all applicable to managing people (even though they’re rarely used in this sphere, possibly because assessing the output of knowledge work is harder than assessing the output of a factory).
First, we’ll look at strategies directly inspired by the three steps of production. Then, we’ll look at some more general strategies.
There are three steps to production in a factory:
To determine the fastest, most cost-effective way to produce products in a factory, manufacturers first determine throughput times (how long it takes to prepare individual product parts) and then offset them (start them at staggered times so they’ll all be completed at the same time and ready for assembly). To do this:
1. Determine the limiting stage, which is making the component that’s most time-consuming, valuable, or sensitive. You’ll organize your workflow around this stage.
2. Determine how long the next-longest stage will take. You’ll time starting this stage so the component is ready at the same time as the limiting stage.
3. Repeat step #2 for all stages.
4. Calculate the assembly time and add the assembly time to the time required for the limiting stage to get the total throughput time.
Three things might complicate your process and assembly:
1. Limited equipment availability. If you don’t have constant access to equipment, you need to factor waiting time into your throughput time. Waiting time could make a different component’s production the limiting step.
2. Offsetting conflicts. If you’re in the middle of one task at the moment you’re supposed to start another, you’ll throw off the timing because you can’t do both things at once. You could ask someone else to help you, but this isn’t dependable. There are three better ways to handle this:
Option #1: Specialization. Assign everyone involved in the process to a single task, and then combine everyone’s work at the end to create multiple final products (instead of everyone individually making their own product). The downside is that specialization is expensive because you need more staff.
Option #2: Purchase more capital equipment. Buy more of whatever tools are needed to complete the limiting step. The downside is the expense of doing this.
Option #3: Increase inventory. Make as many products as you can, so that you’ll have constant access to products so you can supply them as soon as orders come in. The downside is that you’ll have to throw away the extra, which can be expensive.
Choose the most cost-effective option. Do this by studying the pluses and minuses of each factor. You don’t have to use math or a stopwatch; you just need to reflect enough to understand the relationships between everything involved in your production process.
3. Unexpected events. Unforeseen circumstances (such as equipment breakdown) can cause delays and upset the production process. To avoid this, build “slack” (extra time or resources) into the system so that if there’s a delay, you can make up for it.
The easiest way to create slack is to build extra inventory. The best way to balance the cost of building this extra inventory is to keep the inventory at the earliest possible production stage: For instance, keep extra raw materials rather than extra completed products. In every production flow, the material gains value at each step of the process. For example, toast is more valuable than bread, and a three-piece breakfast is more valuable than an egg, toast, and coffee separately. Keeping the inventory at the lowest-value stage saves you money, as if you do have to dispose of the extra inventory for any reason, you’re losing cheap raw materials rather than expensive finished products.
Testing examines the raw materials, parts, or the final product for flaws. First, we’ll look at when you should test, and then we’ll look at the different ways you can carry it out.
As we learned, material gains value as it passes through each production stage. Therefore, you should always try to catch problems at the lowest possible stage to save yourself money.
You should test at three different points during production:
Point #1: Upon Receipt of the Raw Materials
This is also called an incoming or receiving inspection. If you can catch any problems at this early stage, you can avoid contaminating the whole product with one bad component.
If raw material inventory doesn’t meet your standards, you have two choices:
Choice #1: Send it back. If you do this, you won’t be able to build anything until you replenish your supply, which may cost you in lost opportunity.
To avoid getting into this situation, purchase extra raw material inventory. The general rule is that you should have enough to last you the amount of time it would take to replace your inventory.
Choice #2: Use it anyway. If you do this, the quality of some of your final products will be lower.
In manufacturing, the decision is usually based on cost—which is more expensive, an idle factory, or a lower-quality product? (To decide, manufacturers gather managers from the manufacturing, design engineering, and quality assurance departments to discuss.)
However, there is one hard rule: Never use unacceptable material if it could cause your product to completely fail, which is called a reliability problem. Complete failure has unpredictable and potentially dangerous consequences.
Point #2: In Process
These tests check the components while they’re at various stages of completion. There are two types:
Type #1: Functional, which involves destroying a product to check its quality.
Type #2: Process, which involves checking the process that makes the product. This option is generally better because you don’t have to destroy any product.
Point #3: After Assembly
This is also called the final or outgoing quality inspection. The final, assembled product is tested.
Testing costs money for two reasons:
Therefore, to minimize costs, it’s important to carefully consider what type of testing to use. There are two ways to test and you can use either of them at any stage of the process:
1. Gate, in which all material is held at a certain stage until the inspection is complete. If the material passes, it all moves to the next step, and if it fails, it all gets thrown out or returned to an earlier stage.
The disadvantage of this method is that gate inspections take a long time and are expensive. The advantage is that no substandard material makes it to a higher-value stage.
2. Monitoring, in which you take a sample of the material to test while the rest of it moves onto the next stage. If the sample fails the inspection, you can stop the line. Monitoring is the best way to test unless you have reason to believe there’s going to be a big problem.
The disadvantage of this method is that some substandard material might get to a higher-value stage before the sample results come back. The advantage is that it’s less expensive as you’re not stopping the entire production line, which means you can do more of it for the same price as a gate inspection. Therefore, you might actually end up with a higher quality product even though the risk of substandard material sneaking through is greater.
You can also increase the efficiency and cost-effectiveness of testing by testing on a variable schedule—testing according to need. For example, if a factory goes days without finding a problem, inspectors can test less. If problems develop, inspect more often until the quality is the same as it was before reducing inspections.
In the previous chapter, you learned how to increase your productivity by taking inspiration from manufacturing’s production process. In this chapter, we’ll look at some other manufacturing strategies that can aid productivity, as well as strategies for efficient and productive meetings.
Indicators, or measurements, tell manufacturers (or managers) about what’s going in the production process (or administrative process) and provide information about output.
There is a myriad of possible indicators ranging from equipment downtime to profit to customer satisfaction. To determine what your indicators should be, ask yourself what information you want to know first-thing every day to head off potential problems. Effective indicators must:
For example, here are the five most important pieces of information to a factory manager:
Indicators are very helpful in types of work besides manufacturing because they:
People tend to overreact to what they’re paying attention to. (Shortform example: If you’re trying to avoid overstaffing, you might inadvertently reduce your staff so much you don’t have enough people to manage the workload.) To avoid overreacting to an indicator, pay attention to all areas of your business.
There’s a special type of indicator, called a leading indicator, that tells you what might happen in the future. These indicators are useful because they predict problems before they happen, so you can take preemptive action to avoid them. (However, this is harder than it sounds, because you don’t have an existing problem to justify the expense or anxiety of reacting.) Leading indicators must be accurate and believable (or else you’ll be reluctant to act on them).
There are a few leading indicators that are appropriate for most cases. All of them are only visible once you’ve plotted the data on a graph:
Linearity indicator. This indicator is the slope of the line of output over time. Ideally, the line should be straight and ascending (which indicates a consistent output) and hit the desired point of a particular output at a particular time.
If the graph is a saggy line or not a line at all, then output isn’t constant and you’ll have to adjust production in the remaining time to meet the goal. If you continually have concentrated output right before the deadline, you’re probably not using equipment and manpower efficiently and run the risk of having a problem that derails the goal right before the deadline.
Trend indicators. Like the linearity indicator, this indicator shows output over time but also takes quality into account: Output must meet a certain standard. This helps you extrapolate both future output and quality.
Stagger chart. A stagger chart estimates future output numbers a few months in advance and is updated as the real numbers come in. This allows you to see how close the forecasting was to reality, which informs your future predictions (Shortform example: If you consistently predict selling more than you actually sell, start predicting lower). Stagger charts are particularly useful for forecasting economic trends.
| Forecast for: | ||||||
| March | April | May | June | July | ||
| Forecast made in: | March forecast | 20 | 26 | 32 | ||
| April forecast | *21 | 25 | 31 | 29 | ||
| May forecast | *19 | 28 | 28 | 33 | ||
The second strategy is building to forecast, in which manufacturers (or managers) predict how many orders they’re going to get for each product and build enough to satisfy them. The advantage of building to forecast is that customers don’t have to wait for products. (If manufacturers waited until they had orders to start building—referred to as the “build to order” method—customers would have to wait longer for their products.)
Both sales and manufacturing forecast so that both departments are accountable for meeting their predictions, and they use stagger charts to help with accuracy.
The disadvantage of building to forecast is that if the predictions are wrong, manufacturers end up with a bunch of leftover inventory or lack the inventory customers do want. However, this is still preferable to making customers wait.
Delivering built-to-forecast products involves two simultaneous processes with their own timelines. Ideally, both flows finish at the same time:
However much you plan, it’s rare that the two flows line up exactly—sometimes, customers change their minds or there’s a mistake in the manufacturing process—which is where the slack (extra time or resources) you left in the system will come in handy.
Some businesses use both building to order and building to forecast—for example, the breakfast cafe builds breakfast to order (cooks only start cooking after a customer has ordered) but buys raw inventory (purchases eggs, toast, and coffee) based on forecasted demand because it’s not possible to acquire it instantly.
In manufacturing, most workflows are tested and well-worn. Manufacturers don’t waste time coming up with a new way to do something if there’s already a good existing way. Do the same as a manager. However, keep in mind that the value of a process is the logic behind its steps, not the steps themselves, so be open to changes if old processes become illogical.
All manufacturing processes include set-up time, and if you can do all the activities that require the same set-up together, you can save time. This process is called “batching.”
If a factory is already booked up, manufacturing leaders don’t start new projects because otherwise, the overload might cause a bottleneck and materials might have to be aborted at a higher-value stage because they spoil while waiting to move on to the next step.
It’s harder to avoid overloading capacity in a managerial context because you don’t have the same capacity indicators as a factory, but you must have a vague idea of how much time it will take you to do your usual tasks. Say no to new projects you don’t have time for. (Remember that if you’re already fully booked, you can’t get the new project done. Also, remember that your time is finite, so anytime you say yes to something you’re turning down something else.)
Factories try to evenly distribute their workload over time. Managers can benefit from this technique too. The main threat to an even managerial workload is interruptions because they unexpectedly increase workload. You can manage interruptions by taking inspiration from factory techniques:
Make standard (not custom) products. In this context, “products” are answers to questions that interrupt you, which usually come from subordinates or people you influence. Study these questions and prepare widely applicable responses. In addition to saving you time writing original responses every time, you can share these responses with others and delegate their delivery.
Take advantage of indicators. Indicators help you avoid losing time to interruptions by allowing you to do two things:
Encourage other managers to embrace regularity. Because you work with them, your schedules have to mesh.
Batch subordinate interruptions. Don’t physically hide from your subordinates when you don’t want to be interrupted—their questions are legitimate. Instead, schedule a time when you’re available for questions.
In the previous section, we looked at some strategies to make managerial work more efficient. Now, we’ll look at how to make meetings more efficient.
Many managers hate meetings and think they’re a waste of time and a symptom of disorganization. In reality, meetings are just a medium in which you can do activities, and if run properly, they can be an efficient way to carry out the managerial activities that you learned about in the previous part (many of which have to be done in-person).
There are two types of meetings, and if you use your meeting time perfectly, you’ll only ever need to hold the first kind:
The purpose of process-oriented meetings is to exchange expertise and information. These meetings take place according to a regular schedule.
To make these meetings effective, use manufacturing strategies:
There are three subcategories of process-oriented meetings:
The goal of this type of meeting is to exchange information and maintain professional relationships.
Attendees
The only attendees of this kind of meeting are you as a manager and one subordinate.
The subordinate’s job is to:
1. Give you detailed information about her work and concerns.
2. Plan the agenda and give it to you in advance. This is for two reasons:
The agenda should include a discussion of:
3. Take notes on the agenda during the meeting. This is because:
Your job, as a manager, is to:
1. Soak up the information the subordinate provides.
2. Help the subordinate articulate her problems. When you think the subordinate is finished talking about something, ask another question to determine if she thinks she’s finished talking about it. Keep asking questions until you both feel you’ve covered everything.
3. Take notes on the agenda. This is for the same reasons the subordinate should take notes, and additionally, your notes will help you monitor the subordinate as she carries out the actions discussed in the meeting.
Scheduling and Frequency
Schedule the next one-on-one at the end of the current one rather than using a set schedule. Otherwise, a set schedule may conflict with vacation or a commitment. Plan for every one-on-one to take at least an hour to give subordinates time to talk about complicated issues. (If a subordinate knows she only has 15 minutes, she’s not going to bring up anything that seems thorny or delicate.)
There are two factors to consider when determining how often to have one-on-ones:
Setting
Hold the one-on-ones near the subordinate’s work area. This allows you to observe her and learn more about her, such as whether her work area is organized and how often she gets interrupted.
If it’s not possible to meet face-to-face, you can do the one-on-one remotely, but because you can’t watch each other take notes, you should trade notes after the meeting so it’s clear what each person will do.
The goal of the meeting is to facilitate peer interaction, learn the dynamics of your subordinates, and exchange information in both directions.
Attendees
The attendees of a staff meeting are you as the supervisor and all of your subordinates.
Your job, as the supervisor, is to:
The subordinates’ role is to participate in the discussion, which should include talk of:
The goal of this final meeting subcategory is to facilitate learning and connect people—often managers—who don’t have another medium for interaction (for example, they’re too far apart on the organizational chart to be in the same staff meetings). The meeting should include presentations by managers of different groups.
Attendees
The attendees are two managers—the one who organizes the meeting and the one the meeting is directed at (“reviewing manager”), presenters, and the audience.
The organizing manager’s job is to:
For example, at Intel, this role might be held by the divisional marketing manager.
The role of the reviewing manager is to:
For example, at Intel, this role might be held by the general manager of a division.
The role of presenters is to:
The audience’s job is to:
The purpose of these meetings is to solve a problem by making a decision. They’re scheduled in response to a problem’s development. In theory, these meetings should never be called because all problems are preempted in process-oriented meetings. In practice, however, even if things operate well, about 80% of problems will be handled in process-oriented meetings and the remaining in mission-oriented meetings because not everything can be preempted.
Attendees
The attendees are the chairperson (often the person who calls the meeting) and the other people she recruits to help her solve the problem.
The chairperson’s role is to:
Manufacturing strategies work well in a business environment too.
All tasks involve set-up time, so it’s most efficient to do similar tasks all at once than to set-up multiple times. What activities do you do that you could batch?
What leading indicators could you use to predict your future output? What will each indicator tell you?
Think of a project that you monitor. What kind of tests could you do to check its progress?
In the previous two parts, you learned how to improve your managerial output and productivity by adjusting your activities, leverage, and time management to improve the performance of your team. Now, we’ll look at how to improve the performance and output of an entire organization by effectively structuring and wielding control.
As organizations grow, they have more to keep track of and they often benefit from restructuring. There are three organizational structures that you can use to structure your company:
In this centralized structure, individual business units are only responsible for their unit-specific tasks. Any function that they share with another unit (for example, human resources), is handled by a group that handles the shared function for all units.
There are advantages to this structure:
There are also disadvantages:
In this decentralized structure, every individual business unit is responsible for both its unit-specific tasks and all other tasks that come with running a business, such as hiring, purchasing, offices, and so on. Each unit is responsible for its own hiring, purchasing, offices, and so on, and each unit reports to a regional office or the corporate executive office.
There’s only one advantage to this structure, and that’s speed of responsiveness—each unit can quickly respond to local conditions, such as bad weather that slows down delivery at one franchise location. Speed is very important—the point of businesses is to respond to their environment.
The disadvantages of this structure are:
This structure is a combination of mission-oriented and functional. Ideally, it harnesses all the advantages of each system and dispenses with the disadvantages. Companies can and should shift closer to and farther from each pole as they acquire new resources and adapt.
Grove’s Law states that any large organization that has a common business purpose will ultimately move to a hybrid structure. This is because as businesses get bigger, they have more things to keep track of and need the advantages of a functional structure as well as those of the mission-oriented approach.
The major challenges of the hybrid structure are how to allocate resources between business units and how to handle resource-related conflicts. Middle managers are perfectly poised to solve these issues—there are enough of them to manage the workload of allocation (corporate managers aren’t numerous enough) and they’re close to those who need the resources to know how much each unit needs (central allocators are too far away to accurately predict demand).
Another way to ensure the smooth running of large organizations is to have people report to two managers, or for people to be part of more than one organizational chart. We’ll look at both cases below:
Dual reporting is when someone reports to more than one manager, who can be an individual or a group. Typically, one of the managers supervises the technical aspects of the job (for example, how to manufacture a computer) and the other supervises the general aspects (for example, showing up on time).
Dual reporting is advantageous because:
The disadvantage of dual reporting is that organizational charts become more ambiguous, and most people don’t like ambiguity. However, it’s a necessary evil. When an organization gets big and complicated enough, dual reporting is the only way for an organization to function effectively.
Dual reporting shouldn’t create bureaucracy or busywork (work that takes up time but doesn’t produce anything of much value). Keep applying the factory principles: work simplification and regular evaluations of processes to make sure they make common sense.
In a multi-plane organization, a given employee is part of multiple unrelated organizational charts. People don’t have to keep the same role in all planes—they can be a supervisor in one plane and a subordinate in the other. There’s no confusion between roles and they don’t conflict with each other.
This is advantageous because:
1. People can join transitory teams to quickly solve problems.
2. People can participate in more planes.
Often, many of the planar structures are transitory teams—task forces or informal groups that only exist for as long as the problem they’re created to solve.
In the previous two chapters, you learned about the planning and organizational parts of running a company. In the final chapter of this part, you’ll learn about control modes. There’s no universally optimal way to control people’s behavior—the most effective approach always depends on the circumstances.
There are three ways to control your behavior and the behavior of others. We encounter all three modes throughout our lives, including in the workplace:
In this mode, people’s behavior is controlled by price and self-interest. Whoever is buying wants to pay the lowest price, and whoever is selling wants to sell at the highest price possible. None of the parties care about the other—they’re concerned only with their own wants. Therefore, you should use this control method when the parties are mainly motivated by self-interest and the complexity, uncertainty, and ambiguity (CUA) of the circumstances are low (for example, what’s bought and sold has a value in dollars).
No management is needed when using this method because everyone openly follows self-interest.
In this second mode, behavior is controlled by mutually agreed-upon guidelines that outline what each party will do to what standards and who has the right to monitor. Contracts are useful when exchanging goods or services that don’t have a defined value, like the contribution of an individual engineer. Use this method when parties are mainly motivated by group interest and CUA is low.
Management is needed in this method—management helps set and enforce the guidelines.
In this last mode, behavior is controlled by culture—people believe and trust that the whole group shares ways of doing things, values, and goals. Culture is a useful control method when it would be impossible to define dollar values or contracts because the circumstances change quickly or are unclear. Therefore, use this method when the parties are mainly motivated by group interest and CUA is high.
Management’s role is to develop and establish culture, which you can do in two ways:
(Note that when the parties are motivated by self-interest and the CUA is high, no control method works effectively.)
The appropriate behavioral control method depends on your circumstances.
Choose an employee you manage and reflect on one of her tasks. Is she motivated to complete this task by self-interest or group interest? How do you know?
Describe the circumstances she does this task in. What is the complexity, uncertainty, and ambiguity (CUA) of her work situation?
Keeping your above answers in mind, should you use free-market forces, contracts, or culture to control the behavior of this employee? Why?
Throughout this summary, you’ve learned a variety of techniques that will help you become a high output manager. In this appendix, Grove provides a list of activities that are each worth a certain amount of points. Do enough of the activities to earn at least 100 points and you’ll improve your managing game significantly.
For 10 points:
For 20 points:
For 10 points:
For 20 points: