© 2001, 2014 don gray and Dan Starr
Near the Blue Ridge Mountains in North Carolina, not far from where you should think, there is a small town called Mayberry.
Although the main road bypassed the town a few years ago, Mayberry is the same name character of the popular 1960s TV series. It is still a bustling community. Every morning, the commercial branch of highway 52 in the north of the United States enters downtown Maybury from the north. After a week of consultation in the town, we were able to observe the nearest road construction on this route and watch three local residents show their unique management style. Let’s take a look at how these people’s traffic management andsoftware project The general style of management fits (implies).
When the road project in the north of the town closes highway 52, all traffic entering the town from the North must bypass highway 52 to the west of the town, and then enter the urban area of key street. Unfortunately, this means that the traffic will have to turn left into key street and cross the rather busy east-west traffic (see Figure 1).
The town council was worried that the traffic that hurriedly waited for a left turn into key street in the morning would go all the way south along the ramp and extend to highway 52 itself. Since the speed limit on highway 52 is 65 mph, it may lead to serious accidents. Therefore, the parliament decided to send a policeman and one or two rescue team volunteers at the intersection to ensure that the traffic on the ramp would not be congested.
Three management methods
As a duty officer, the duty officer (we call him Barney) arrived at the scene on Monday and quickly determined the situation. What he thought was needed was traffic lights at the intersection of key street and the ramp. Because it took several months to approve a light, he decided to operate it as an “artificial traffic light” to guide the traffic manually. There are turns in each direction: westbound key Street (including turning left into southbound ramp), then eastbound key Street (including turning right into southbound ramp), and then downhill (turning into key street in any way). Barney’s plan is actually not that good. Two way traffic on key Street stalled. When Barney made several cars turn left to key street, almost all the traffic was blocked on highway 52.
On Tuesday, a volunteer for the rescue team (a helpful local woman called aunt BEA) said she knew how to manage the situation. She believes that as long as drivers do not have to cross each other’s roads, traffic can take care of themselves. Therefore, she let the traffic flow drive in both directions on key street and let people turn right and right on the ramp. When someone has to turn left, she stops the other lanes and lets them go. Aunt bee’s method worked better than Barney’s (at least no one was rude to her), but the congestion was still much more than we expected, and it remained at the end of peak hours.
On Wednesday, Sheriff Andy showed up with a lawn chair and a lemonade thermos. He put the lawn chair in a cool place, where he could see a good section of the intersection and downhill road, and then sat down to drink lemonade. When the traffic began to jam on the ramp, he got up and stopped the traffic on key street, leaving the ramp empty. Then he went back to the chair. Besides, Andy seems to have almost nothingdo。 Although he clearly did not take any action, the intersection still seems to be effective. People stay calm and relaxed. The driver turns to the right, creates rest time for others, turns to the left, and everything works like before anyone else. The situation is better, but a little better.
Wearing the consultant’s hat, we realized that we had just witnessed three unique management styles: Barney’smicrocosmicMicro management, aunt bee’sMotherlyMother management and Andy’sExpert styleMaster management. Because these styles are also very common in software project management. Let’s study each in more detail and understand what can be applied to our own software projects.
Each manager shapes his own style by different assumptionsSomeone who manages, and assumptionsManager’s role。 These assumptions determine how they conduct key management activities. At Weinberg’sSoftware quality managementIn a book. 1: Systematic thinking, Weinberg highlighted five key activities:
- understandProblems to be solved，
- observationThe actual behavior of the managed personnel,
- useRules and process modelTo determine what to do next, and
- take actionGuide the group towards the goal.
These activities together constitute a feedback system to “guide” the project team. The way they are implemented (i.e. the way managers define problems, how managers plan, what observations are made, what rules are followed, and how corrective actions are taken) makes everything different – determine where the team is going, how team members feel about the entire software project, and how satisfactory the final result will be.
Barney implemented micro management, which is based on the following assumption: managers must ensure that everything is completed. Most micro managers do not deliberately intervene for the purpose of control. They just operate on the assumption that if they don’t, they won’t be able to do it. Micro managers also tend to make relevant assumptions, that is, the managed will follow their instructions. Just right, no more, no less.
These assumptions are better than people’s description of machines. Indeed, when Barney said that we needed a “traffic light”, he described the situation where managers and managed people were more mechanical than people. Maybe that’s why so many excellent programmers become micro managers when they get their first promotion – they are just “programming” the “biological robots” working for them!
Using Weinberg’s model, we can see how Barney’s hypothesis defines his view of key management activities:
- takeProblems to be solvedIs to personally ensure that everything is going on in an orderly manner.
- NextplanIs to let Barney do everything himself. He will personally direct the movement of each car. This means that the plan must be simple enough so that he can control its implementation at any time.
- Even with a simple plan, Barney was too busy directing traffic toObservedToo many things. Standing in the middle of the intersection, when the traffic began to fall back to highway 52, he didn’t look at the ramp in the right position.
- Even though he has better observation, he is manager centeredProcess modelHe is still not allowed to do too many things. The basic assumption is that he has personal responsibility for every car crossing the intersection, which means he can’t delegate too much — he can’t expect drivers to do anything other than what he wants to tell them.
- Barney’smannerVery limited. Because he has to control every car, he can’t leave the middle of the intersection. Finally, he had no choice but to try what he had done – waving his arms to people more madly in the hope that they could spend it faster.
Because managers must make (or at least approve) all decisions, only one thing happens at a time, while everything else waits in line for the turn. When simplicity, centralized information and supervision change from virtue to disadvantage, it will create bottlenecks affecting project planning and implementation.
Simplicity because the whole project plan must always be under the control of the manager, the plan must be simple enough for one person to fully understand it. This sets an upper limit on the complexity of the project – if the problem to be solved exceeds this limit, the manager must simplify it in some way (for example, only one direction of traffic is allowed at a time). In micro management projects, the serialization of such activities is also a common simplification, and wastes energy and time. When serialization is not enough, managers may begin to exclude “unnecessary” activities from the project plan. Micromanagement is notorious for over simplification, so that their software project plans may miss what is necessary for the successful launch of products.
Centralized information since the manager is the only one who can make decisions, it is important to obtain a large amount of high-quality information about project implementation. Unfortunately, the only observation allowed iscontrollerObservations made in the project plan, but the manager is too busy making every decision to actually observe anything. Therefore, in practice, micro managers often turn a blind eye and make decisions (hasty decisions) when there is little or no actual information.
Supervision requires explicit approval of each operation, which increases the time required to complete the task. Therefore, micro management is often inefficient, and many people are waiting for managers to tell them what to do next. Bottleneck managers are a key structural problem. This practice can also lead to personnel problems, such as active inhibition. The manager’s assumption means that the managed personnel have no contribution other than the functions defined by the manager. What if workers want to do something that doesn’t comply with the rules – because they see problems with better methods or plans? forget it. Micro managers will not allow it to happen. For those who are micromanaged, this can lead to short tempers and long days.
Most people don’t like this management style. Some people will make some rigid, mechanical compliance response and dutifully wait for the next set of instructions from the manager. Others may choose some form of subtle rebellion, such as “trying to rule” – follow the manager’s instructions even if they are clearly the secret of failure. Others will rebel more openly, using managers’ constant distraction to escape everything. Micro, these responses to micro management often establish a positive feedback loop, which strengthens the assumptions of micro managers and leads to more micro management. Micro managers are often busy. (the more attention to detail, the busier)
So, was micro management ever appropriate? Of course, if the problem to be solved is small enough, a manager can really understand the whole project plan, and the personnel engaged in work are willing to follow every command of the manager. Although this happens sometimes, it is not common in the software world.
A common reason for micromanagement is that newly promoted, technically qualified managers rush to help troubled employees or rescue specific parts of software projects. This forms an interdependent driving force, in which managers become life-saving straw, but employees become helpless. This ensures that the next time a problem occurs, the manager will intervene again, and so on, until something breaks the pattern.
Although micro management projects can (and often do) lead to successful product launches, it is more about managing them than it is about it. There should be a more effective and easier way to deal with this situation.
Aunt bea chose a more friendly and gentle style, which we call maternal management. Allow the driver to do something by herself and help them when she thinks she needs help. But her basic assumption is still very close to Barney’s assumption: the person under management may be able to do some routine things without being told, but all important decisions – especially when there is some form of competition – are still firmly under her control.
If micro managers manage people as machines, aunt managers will treat them as children. The managed person can do some routine things, but still needs to be protected from any potential danger. Like micro managers, aunt managers are not necessarily malicious or in urgent need of control. Aunt bea doesn’t need to control the driver at all. She just knew that without her help, they would not be able to make major decisions. She could hardly imagine the gap caused by one person turning left into another turning right, because she could not see who was controlling the relationship, and she knew that the two drivers certainly could not coordinate them without someone.
Aunt BEA’s mother hypothesis determines her view of key management activities:
- takeProblems to be solvedIt’s like “taking care of people who have to cross other businesses.” Like Barney, she sees things from a personal perspective. It’s her problem, not the driver’s problem.
- Because Aunt bea sees the driver as someone who can do something by herself, sheplanThan Barney’splanBe strict. She can allow at least some routine things to happen in parallel, but in special cases, she will have complete control of everything, which means returning to serial execution.
- Aunt BEA’s decentralization plan requires more complex than Barney’sobservation。 She must observe the situations that need her help, especially the left turn. Attention, sheNo,Observe whether people have trouble turning left. Her basic assumption is that turning left is a help signal. Like Barney, she spent her time in the middle of the intersection, and she couldn’t see the ramp well from this angle.
- Because the aunt manager believes that the person under management cannot handle any form of dispute or conflict, aunt BEA’sprocess model She is required to solve these problems herself. Therefore, her response to any abnormal situation is to stop the traffic and help solve the problem.
- Like Barney, aunt bee’sactionVery limited, partly because she needs to be in a controlled position in the middle of the intersection.
Like micro management, mother management can work when the basic assumptions of parent management are correct and the problems and solutions are not too complex. The trouble is that most software development is not this model, and most development is unconventional and needs to solve many conflicts. Interface, partition, decomposition, protocol – these are what aunt manager thinks of as “left turn”. He must personally ensure that everyone plays well. This creates a structural problem similar to micro management. Similar to micro management, but also different. Because some work can be carried out independently under the management of the mother, managers have no difficulty compared with micro management.
However, because the process is still highly manager centered, the actual workload that can be completed in parallel is usually less than expected. We finally got a very effective process: almost parallel, relatively observed and very close to giving workers independent responsibility:
Parallelism (almost) only predefined “normal” things can occur in parallel. As long as the traffic goes straight ahead or turns right, aunt BEA’s plan seems to work. But she couldn’t predict how many people would want to turn left. When many people began to turn left, her plan collapsed. In the same way, the actual performance of software projects managed by Aunt largely depends on how much development is really needed without interaction or conflict resolution. If there are more “exceptions” than expected, many developers working in parallel according to the project plan may sit in their hands and wait for the manager to make a decision. This can make in theoryparallelIn practice, the project plan becomesserial of
Nanny for the managed people, aunt management is lighter than micro management, because the “mother” allows her “children” to do something by themselves. Individual developers can move forward as long as they don’t deviate from the process or fall into conflict. However, if non-standard behavior is first indicated, the whole process will stop until the manager decides what to do. Managers must deal with all really important decisions, which stifles the individual’s contribution to solving the overall problem, almost as ineffective as micro management. There is a big change here – a manager who sees employees as teenagers is less open to control than those who see employees as children. However, most people in the software business have a college degree, if we want to find a better way thanmicrocosmicandmaternalFor a more effective style, you must start by changing the basic assumptions. Barney manages people as machines to be programmed. BEA thinks they are children in need of help. Now, let’s see what happens when Andy sees them as adults.
Andy’s approach doesn’t look like “management” at first. He just sat in his chair, drinking lemonade and watching the traffic on the ramp of highway 52. When the ramp began to be seriously congested, he slipped into the intersection, stopped the traffic on key street and cleared the ramp; Then he returned to his position. He seems to do less “work” than Barney or aunt bee, but the traffic is smoother. We call Andy’s styleexpertManagement – because of our three traffic controllers, only he is the real master of management.
The key to Andy’s management style lies in his basic assumption: drivers are adults who can take care of themselves most of the time, and his role as a manager is to support these capable adults so that they can do it safely through the intersection. This is completely different from Barney and aunt BEA’s hypothesis. Andy feels safe enough about his abilities and driver skills so he can move himself out of the work center.
Since Andy does not place himself at the center of management tasks, he can be more flexible and effective in key management activities:
- Andy thinksThe problem to be solved isCross the intersection efficiently and safely. He also realized that most of the time this intersection did not need any help. People turn here every day without any supervision. What makes some management and intervention necessary here? Detour increases the traffic on the ramp of No. 52 expressway, which may sometimes lead to traffic jam on the ramp to the expressway and cause potential safety hazards. Please note the difference – Barney and aunt Bobby define problems according to what they have to do, while Andy defines problems according to the results, regardless of who really “finished the work”. By doing so, Andy positions himself as a system that observes and “guides” work, not as a person engaged in the job. (translator’s note: focus on skills vs. focus on results)
- By understanding the actual problems to be solved, Andy can provide solutions for themDevelop effective plans。 The driver is responsible for crossing the intersection himself. Andy and his “management team” will monitor the ramp and ensure that it is cleared when (and whether) the congestion is far away and poses a safety hazard. Although Barney may accuse Andy of not having too many plans, the fact is that Andy’s simple plan actually allows some very complex things to happen. Because he didn’t try to control the low-level actions of the driver, Andy’s plan entrusted the management to each driver. In this way, they can run in parallel, and they wait – drivers waiting to turn left take advantage of the traffic gap created by drivers turning right.
- Now Andy has both a problem statement and a plan. Andy can determine what he needs to doobservation。 In order to prevent traffic congestion and return to highway 52, he must pay attention to the ramp – not the intersection. So he put himself on the side where he could see the ramp. This is another important difference in Andy’s style. Standing in the middle of the intersection, Barney and aunt bea are absorbing a lot of information – most of which has nothing to do with solving practical problems. They are not in the right place to make really important observations. Of course, Andy didn’t ignore what happened at the intersection, but he didn’t take the intersection as his main focus.
- Andy’s management style uses twoprocess model 。 First, if the traffic jams on the ramp, stop the traffic flow on key street and let the vehicles on the ramp go first. Secondly, if something blocks the intersection, please remove it immediately. For the rest of the time, Andy’s process model says “let the driver take care of himself”.
Both models are more subtle than they seem. The first model allows Andy to make some fine-tuning in the morning. How far is the ramp from the traffic distance “too far”? At first, he took a conservative approach, clearing the ramp when it was half the way up the highway. Later, after observing how quickly the traffic on key street could stop to clear the ramp, he changed the definition of “too far” to three-quarters of the distance on the ramp. This means that less intervention is needed, because usually, the traffic will rise to three-quarters of the congestion point and then fall back on its own.
The second model contains flexible definitions of trigger actions. Andy is looking for symptoms that may have multiple root causes. If something blocks the intersection (for example, the driver is too timid to turn left), Andy’s model will deal with it.
- Finally, Andy takes the “open”get some actionMuch less than Barney or aunt. Most of the time, he seems to do nothing. However, when action was needed, he knew which action was appropriate and effective. But it is wrong to say that Andy’s action is simpler than Barney’s or aunt’s. In fact, his infrequent intervention requiresmoreskill. After all, Barney and aunt bea aunt have stood in the middle of the intersection and attracted the driver’s full attention. Andy had to enter an intersection full of moving vehicles to attract the driver’s attention, temporarily interrupt their self-management, let the driver execute his instructions, and finally re-establish the self-management system. This is a task that requires some skills.
Like the other two styles we discussed, mastery management works when its basic assumptions are valid. In software development, if the management personnel are skilled, competent and educated adults, these assumptions are usuallycorrect。 Therefore, skilled management can solve the structural and behavioral problems we see in micro and aunt Management:
The inherent authorization of the plan means that most disputes and minor conflicts can be solved without the intervention of managers, so people don’t need to wait for the orders of managers most of the time. When the problem really needs the attention of the manager, the problem does not have to wait in line after a series of minor conflicts.
Support for parallel activities means that skilled management can handle projects that are too complex to be understood by a single manager, and most software projects fall into this category.
Because the managed people are also assigned to self-management, they can put forward ideas that micro managers or aunt managers may miss.
Expert management involves managementprojectinstead ofpersonal。 In most cases, workers are free to choose their own methods within some basic guidelines (for example, driving on the right road or using the company’s standard toolset). This allows creative energy to find ways to “beat the system” rather than creating profitable products.
In short, expert managers like Andy observe and manipulate a system. If the problem is well understood, the plan is appropriate, and the person working is competent, the controller usually doesn’t need to do a lot of things. Unlike micro managers and aunt managers, skilled managers spend most of their time observing and thinking rather than crazy activities. But don’t be fooled – when Andy sits in a chair and drinks lemonade, he controls the situation more effectively than Barney or his aunt.
If expert management is so good, why don’t we see it often? Because it is disturbing in some ways, especially for managers:
Appearance can be deceptive. Projects managed by experts often give the impression of confusion. When Andy manages the intersection, the traffic turns in all directions, which is disturbing compared with Barney’s neat and orderly behavior when he is responsible. However, in Andy’s seemingly chaotic management style, more traffic passed the intersection and was safer. Many software projects already look chaotic. Will expert management make them more so? We doubt this; We suspect that much of the apparent confusion in software development comes from resistance to micro and aunt management.
Expert managers need different mindsets. Most people willAdministrationAndpowerThe word. However, the shift from micro management to expert management involves abandoning many of the apparent power and authority of management positions and handing them over to staff. According to writer Barry Oshry (in Weinberg’s book《Becoming a technology leader(quoted), senior managers have stronger power if we define power as “to enhance the ability of the system to develop and act in the environment”.
Measuring important factors in some organizations (especially those based on micro management), expert managers may be difficult to promote. After all, compared with the surrounding micro managers and aunt managers, you will not make much visible changesAdministration, and it is easy for micro managers who make promotion decisions to conclude that although you “do nothing”, the project succeeded, but not because of your management.
But expert management also pays off. Expert managers usually don’t have to work like micro managers and aunt managers. As a senior manager, you are unlikely to arrive at the office at 3 a.m. to try to solve another trivial problem. When the project team says “we did it ourselves”, you will know that you are indeed an effective leader and will be satisfied with it.
Micro management, aunt management and expert management
The best way to determine management style is to ask questions and observe what is happening.
Are the people who report to you scattered like leaves in the wind? Do you think they are fulfilling the legal provisions rather than the spirit? When a problem occurs, do you jump in and start coding? If so, you may have micro management.
Do you organize workflows to reduce interaction and make things go smoothly in your team? Do you step in and try to fit everyone? In the tight mode, do you revert to micro management?
Do you spend a lot of time observing what is happening, considering the impact that the event will have on your team and project, and what you plan to do? If so, you may be an expert manager.
If you want to change your management style, you need to consider some important issues. First, how do you have your current management style? For most of us, the way we manage will be influenced by the people who manage us and our environment. Recognizing these impacts and the limitations of your current work situation may help you decide whether to adopt the new model. It’s also important to check your stylefeel。 If you are satisfied with the status quo, you may not need to make changes. However, if you feel overworked and seem to be fighting the fire all the time, maybe you should make some changes.
Original text: https://www.donaldegray.com/m…
Finally, what do you want to do? We see that Barney (micro management), bea (aunt Management) and Andy (expert management) have influenced their unique response to the “immediate problem”, and the same is true for you. Once you know what you want to do, you can create and implement plans to achieve your goals and keep traffic running smoothly.
Welcome to leave a message to discuss. What style of manager are you (not necessarily the management, anyone can reflect)? Do you feel tired every day, just like fighting a fire? What is the situation and current situation around you? What is important and how to observe from a reasonable position? What action (attempt)?
This article started on Bob Jiang’s blog. For reprint, please contact Bob Jiang