Geekly Articles each Day
📜 ⬆️ ⬇️

10 skills of "effective" managers


Author: Evgeny Efimov
QA Lead, DataArt

Sailors, airplanes and corporations



There is the notion of "systematic error of the survivor." It is found in the books of Taleb “The Black Swan. The Impact of the Highly Improbable ”(“ Black Swan. Under the Sign of Unpredictability ”and“ Fooled by Life and the Markets ”). He describes a beautiful picture, ancient Greek sailors pray to the gods during the storm, the gods like the way they pray, and therefore the sailors are saved. The storm turns into calm or a weak wind, the ship sails to the shore, and everyone remains intact. as we managed to swim, they also say: "We pray to the gods and always swim." that we cannot ask the sailors who drowned. It is possible that they, too, prayed, but did not make it. And in general, it wasn’t that, because, at that time, almost all the sailors were religious, assume that everyone prayed. And there is a possibility that it does not work.
The second story is about the mathematics and statistics of Abraham Wald and about the planes. The action takes place during the Second World War. British bombers return to base. The task is to understand where and how to strengthen the aircraft so that they return as much as possible and fall as little as possible. They look at the planes that have returned, and they say that it is necessary to strengthen the parts damaged by the hits. Since they are shot at all the time, we will strengthen them, and all the bullets will fly back to the enemy plane and kill everyone there.

Wald made him think that all the tails and wings had been shot, but the plane had returned, which meant that we should not take into account precisely those parts. He said that it was better to look at parts of the aircraft that were not hit by bullets. Because airplanes that were damaged elsewhere, did not return. Thus, we should strengthen not the wings, but what is around, for example, the propeller blades or the cockpit, because if they get there, the plane will not return at all. The basic idea was to study not only the planes that had returned, but also the wreckage of those planes that had fallen, in order to try to understand the causes of the fall and then find the damage that they could not continue flying. And, in accordance with the information received, strengthen these particular parts.
')
Stephen Levitt has written an interesting book, Phryconomic. He selected the standard books from the “10 Steps to Success” series and analyzed two books about corporations that used innovations that led them to success. These books were conventionally written in 2000, and corporations, using these developments, won the rest of the company. When Levit analyzed them, it turned out that three-quarters of these corporations had closed. They turned out to be unprofitable, because the external environment has changed, and they have not changed the approach to success. Accordingly, all these success stories do not always work for everyone.

There is such a thing as a “systematic error of a survivor” when we collect statistics where it is easier to find it. From those who succeeded, from those who survived, and we consider only her. And we do not take into account the statistics of those who have failed. All these factors can greatly distort our reality and affect how we will make a decision. Sometimes there is much more interesting information for those who did not survive.

Projects and conveyors



In which case can we collect statistics from successful people who have succeeded? In that case when we work with the conveyor. When a certain algorithm of actions is guaranteed to lead us to a specific result. We first boil the water, then throw dumplings into it, wait, turn off the water and eat. Strongly deviate from this algorithm does not make sense. It is unlikely that any external conditions can interfere. Therefore, once having received some recipe of a dish, a training plan (that depends little on external circumstances), we can use them. This is just the difference between projects and pipelines.
There is some kind of conveyor thing, we came up with an algorithm, and it works fine. We can give the robots, but the project has a completely different essence. The project is something that is done from scratch every time, something that has not been done before. Therefore, our standard algorithms can break at any time. Something may really not work. And one of the important elements is dependence on the external environment. If we have a more or less closed system that does not depend on the external environment, we can algorithmize it. In IT projects there are many processes, solutions, tasks that strongly depend on the external environment, on the customer, on the market, on the team, on our own company, etc. And, if this dependence on external circumstances greatly outweighs what system is going on inside, the algorithm will not help. Because to predict all these external circumstances is quite difficult.

How to use success stories



But you should not be completely upset about this, because all success stories can also be used. It is important to understand how it all ended. There were firms that had know-how, but then they died. It is good to know that they did not succeed, and at what point their success story algorithm stopped working for them, and how it all ended.
It is important to know the full context. When someone says that he has done something, we need to understand what was happening around that moment. For example, we go through interviews and ask a friend: “Tell me please, how did you go, how did it all go?” He replies that he came and just said that I am Bob Pupkin and took him immediately. You think about how wonderful it is, you just say that you are Vasya Pupkin - and you are hired! But in fact, it may turn out that the name of the director is also Pupkin, so this is all that was needed. And for us this option is not suitable.
It is important to understand the transition points when external circumstances have changed and, accordingly, at what point the algorithms either worked or did not work. If we can recreate the same environment, the same conditions, most likely, will also be able to use these success stories. If we fail to recreate it, or if we understand that it has changed, and everything has become bad, and we have changed to the same, it means that we must refuse and look for some other ways.
It is important to listen to the stories of competitors and find out what they were doing. Perhaps they did the same. Read the stories of rich people or large corporations who have failed. There are general principles that in one case lead to success, in the other they do not. We can say that these particular things do not really affect the success.
And it's good to go to the meta level. Do not just say that if you always wear a hat, you will never catch a cold. It is necessary to deduce the meta-rule: if you always dress according to the weather, the chances of catching cold are greatly reduced.

How to use failure stories



How to use the history of failure, the history of someone's file. It seems that everything is simple - do not do it the way the other person did. Not really. With the stories of failure all the same. I deceived you at the very beginning, saying that failure stories would be more useful. They will be really useful, but you still have to think about the context, about the competitors. And display the history of failure to some meta-level. Trying to make them not direct instructions, but some rules that are caused by the external environment.

Top 10 mistakes


I will tell you about the top 10 mistakes that I made.

1 - I seem to be doing nothing



When I was a quality manager, no matter how much managerial work I had, I could always devote a part of my time to engineering. There were days when all eight hours were spent only on various calls, communication with the customer, on strategies and planning. And that's all. And if these actions took two hours, then I spent another six on some kind of engineering. I have always been busy and felt like a good and efficient employee who does not just eat up the company's money. When I became a project manager, it turned out that sometimes I work 12, 13, 15 hours - because I have a lot of things to do, especially at project starts. And there are days when I phoned two times - that's all, and there is no more business. It starts to seem like I'm not doing anything and not loaded enough. I'll take myself another project. As a result, I took four projects at the same time, and it was a collapse. The manager’s download is wave-like, and if you have a long low load period, you like it at first, of course. A couple of days you feel cool. And then you think: “No, well, it is necessary, it is necessary to occupy myself with something, I will take more projects”.
And then you have a recession load changes on the rise, and specifically now you have 20 hours of work for today, although it is only eight hours. You need to clearly understand that if you periodically do not have work - this is normal. If you are involved in management, if you are doing well and well, then from time to time you will have pieces of time when you are not doing anything. They are very afraid and worried, so it’s not worth it, when you become a little higher, you can fill in these pieces with different analytical work, development of relations with the customer, account management. And at the very beginning you will have relatively loose pieces. A meeting is also a job, if you talk all day, it is also a job.
If you have not done anything at all, there are no artifacts of your activity, this is also normal. Because calling up all day and talking to everyone is also hard. On the one hand, it turns out that you didn’t do anything like that, but you still get very tired. And one more important thing: it makes sense to manage yourself. You are your own manager and you set tasks not only for the team, but also for yourself. And this too is part of your work.

2 - PM bottleneck



I became a bottleneck when it seemed to me that I was not doing much, and I needed to take on more of everything. Since I have free time, why not help Business Intelligence communicate with the customer? I begin to participate in this process! Why is there a QA-manager communicating directly with the customer? I will enter this communication! And so on. And it turned out to be a link in all communications — everything that happened inside the project passed through me. On the one hand, this is quite useful, the project should be aware of everything that is happening in the project, but then it turns out that, for example, if you are sick, tired, there is an urgent work that no one except you can do, you Substitute the whole project. Because everyone asks: “Where is Vasya? We have such a schedule that we hand over the report to Vasya, and he sends them to the customer! Do we now have to send it to the customer ourselves, and it will work? ”In short, a bad thing, try not to do so. Delegate authority at a time when it will be important.

3 - Interesting tasks



I had a lot of tasks, and most of them were pretty dull, but some were incredibly interesting. At first I took interesting tasks, and then worked sad, and that, in general, is not very correct. The extent to which you are interested in the task, and in what priority and in what sequence to perform the tasks are completely different matrices that are not connected at all.
Here you can advise to closely monitor the priority of tasks, but this is not a panacea either, because there are different priorities for the team, for you, for the company and for the customer’s business. Ideally, you need to maneuver between these different priorities, but in practice the business, for example, is not always aware of its priorities. He says that this is super important, and those things about which he does not think, he takes it for granted at all. And he says that it is very important for him that we could do, conditionally, upload pictures. And in fact, it is not the loading that is important, but that, first of all, all the likes are able to set. But it seems so obvious to him that likes are more important than anything in the world, that in the end we don’t do likes, but make pictures for him. They will be beautiful and everything will be fine, we bother with all this. And then we hear: “Where are my likes? Why did I do all this? ”
And one more thing that I learned for myself through the long nights of management: absolutely everyone is crap.

4 - Bother everyone



The most terrible thing when you think that this guy is super cool, he always does everything right, and he suddenly messes up. You think that this guy is never mistaken, but the truth is that, because everyone is fake, and he will do something wrong at some point. You thought it was a concrete wall, but it turned out the opposite. First of all, because there are no risk-free areas, even those areas that you regard as absolutely calm and risk-free, where there is nothing to run into, can explode. For example, if you misunderstood the requirements, if what you gave it to a beginner developer, who basically did everything very badly and not only did everything badly in his field, he also undermined everything nearby, and it also collapsed.
Explode can everywhere. Seniors bummed, customers bummed. By the way, the customers do it very well, because you agreed, he said that he didn’t need likes and even wrote a letter in the letter, but he really didn’t mean it. At the end of the project, he cannot accept it, because he cannot unload it, because he promised huskies to investors, and no one needs projects without likes. And he honestly says: “Yes guys, I was also wrong, but, nevertheless, we cannot do anything, that is, you are free from responsibility, but the final goal — to carry out the project so that everyone is happy — is not achieved” . And this, too, must be always ready. You need to be able to help the customer, very carefully explain to him whether he is sure now that we should do just this way. If you just switched to project management, you will mess up the most. You will be like the Achilles heel and this is also normal. It is important not to mess too much, it is important not to make too serious mistakes.
Mikhail Zavileisky, one of the directors of DataArt, who has a lot of good speeches ( they can be watched ), said that when he was in business school, there were many business games that work on different cases and solutions that are close to reality. He said that the secret of success in all these games is not to come up with the most effective strategy or not to come up with the most error-free strategy, reducing the risks to zero, but simply not to be mistaken too much.
Any strategy that you come up with does not save you from mistakes. It should be more careful in your decisions, then you are likely to win and get a plus. It is important not to reflect and not to suffer, to try to carry out everything in a rational manner. Ask yourself if I did not do it the way I should do everything right next time.

5 - Distortion Discount



The next thing that follows, from the fact that everyone messes up, is that there will always be distortions. Will lie as estimates, and plans. This does not mean that you need to abandon estimates and plans, do not plan anything, do not evaluate anything, etc. You should always remember and take into account distortions. Another important thing - all these distortions accumulate. That is, if your programmer has rated 10 tasks each hour, you, as a cunning manager, said that it would take 20 hours (10 hours) and the programmer did the first task in two and a half hours, this means that he will do it not in 11.5 hours, but in 25 hours. This means that these distortions will continue to accumulate. Therefore, you should always make amendments in the course of the project and watch what happens next.
There is one more thing: an amendment must always be made in plus and never in need - in a minus. Because programmers are used to having managers inflate their grades. In order for the manager to make sure that the programmers are great, that they can appreciate the time well, they can say that it will take three hours and do it in an hour. But in fact, this task was worth 15 minutes. The fact that people take at the beginning of the project, for example, the construction of architecture, is quite a lot of creativity and quite a few restrictions. You really can do it quickly, easily and beautifully. And you will have to redo everything later, in the middle of the project, when everything will be integrated, but now it will turn out - you’ve determined that you’ll work on the architecture for a week. The architecture is done in three days, and you start thinking that you can pass the entire project much faster. You should never give a lower rating.

6 - Micromanagement



The situation becomes even worse when you start to engage in micromanagement. This problem is also related to the previous examples. It seems to you that you are not sufficiently busy, then you understand that everyone has screwed up, that all their tasks are not doing on time and wrong. And the only thing you can do is to approach everyone every five minutes and ask when everything is ready. Already done? Checked out? Hate you will be on the second day. This is bad for you and the team. Because you think: “So, I asked this, and I asked this, what, god, 15 minutes have passed?”
The root of evil is in its own incompetence. If you are super-confident in your managerial abilities and in what happens in a project, you will never be involved in micromanagement. When you do not understand what is happening around you, you have a feeling of complete loss of control, you are trying to follow everything through micromanagement. From it will be worse for everyone, including the customer, because you will transmit this micromanagement to him, telling him every 15 minutes what is happening, that everything is in order, we have no problems. And the customer will then wait for this all the time.
There is a chance to build a perverted system (sick system). She is not inclined to produce something, but is aimed at preventing people from leaving it, no matter how bad it is inside. Usually it works through guilt and responsibility, constant hard work and promises of a good life.

7 - First a case - then reports



The next interesting point. You have a job and hell at work, you need to add more people, you need to fix everything urgently. You need to constantly offer the customer the best versions, show that you have progress. And you tell everyone: “Let's work!” And forget about such things as reports, JIRA, tasks, Burndown and making adjustments to the plan in accordance with what is happening now.
You are not a manager, but as a leader you lead your team into battle. Customize and cheer the team. Tell everyone how important it is to finish everything now. Constantly rush everyone: “What did you do? Let's check. You negotiate with the engineers who deploy the servers to fill in everything and do everything quickly. In general, you are constantly solving some problems. Reports? And so it is clear that everyone works. All boils. The team turns on emergency mode and gives incredible performance. The team spirit is incredibly high and it seems that now we are catching all the problems.
In fact, everything is a little different. Everything becomes very bad because transparency is lost. And the customer does not understand what is happening here. He sees everyone running. And where is his product and when everything is ready, he is not very clear.
Tensions are growing both inside the team and outside, especially from the customer, if he does not see transparent reporting, trust is lost, he begins to put more pressure on the team. He puts pressure on the team, the team, and so under pressure, in general, nothing good will come of it.
As a result, your planning horizon collapses because you no longer understand what is happening now. You think in very small categories of specific tasks, tasks and problems that need to be solved now so that everything becomes good. And what product should be in a week or a month, you no longer see, the planning horizon collapses. Although working in this mode without reporting and plans, constantly helping everyone, is very pleasant, it is better not to do so.

8 - collapsing horizon



Supplement about the collapsing horizon. So, you have a bias towards tactics and towards short-term goals. This is when some time trouble in the project occurs. And if you have carefully done all the previous, most likely, it will happen to you. First of all, you may not see something very big and important, about what you forgot from the very beginning, because at the previous step you already forgot about reporting and suddenly it turns out that you put on likes, you had pictures from the very beginning, but Actually, it would be nice to stream video as well. But this is a task for a month, we had it much later, and we started doing it just now, and began to lose deadlines. We have done everything, we have advanced in terms, we have done great work, we have won all, but we have forgotten about this important task. Because we have not seen before that we need to come.
When we lose a course, the nervousness in the team may increase. Everyone can have fun doing current tasks and getting new ones. And the manager ceases to understand what is happening, because he does not see what should be ahead. The customer begins to get nervous, because he does not understand at all what will happen next.

9 - Aid Balance



One more thing that I didn’t really understand is where your responsibility ends from below and above. Should I give each engineer a personal call and show detailed instructions, how should he do all the reports, or from his words, write down everything he does, and then put it into a report? Or can I ask him, describe everything that he did, and only bring him down? Or I can assign my person to each week, who will send these reports, and not do it at all, but only view them. And how is my responsibility limited from above? How responsible am I for the entire project? For budgets? Should I come to some more senior manager and say that we are not coping and we need more people. Or is it just my problem, and I have to decide myself what to do with the fact that we are not coping. It is important to find these areas of responsibility on both sides and clearly speak the boundaries.
You are the project manager, there is some kind of leadership above you, especially when it comes to communication with the customer. It is necessary to decide whether we are ready to tell the customer that we will enter any feature or not - any feature is probably not ready, only a feature that takes less than four hours conditionally. If he asks for a feature once a month, you can do it without agreement with someone.
It is necessary to agree on these boundaries from the very beginning. My main mistake was that by requesting help, I completely delegated responsibility, in the end, my manager at some point began to do my work, instead of helping me with resources or expertise.

10 - Anticipate expectations



Never promise more than you can do. Always try to promise less and do more. Especially cool when you give a customer a conditional gift. Without any problems, you agree to make some additional feature for the customer, if you are sure that it is small and it is simple (even better, if it is already done).
For example, you, when a customer asks about the dates, promise that everything will be ready tomorrow. You could not do this, but you have already done it, and if tomorrow it is not ready, it will be much more upset than if you would say to him at this moment, now it will not work, it’s better to transfer it to the next sprint and there all is well done. And if you promised and did not fulfill, it will look much worse. The customer will be upset, he may have already promised everyone that it will be done, and then once and not received anything.
Rule No. 1 - “talk less and do more” - must always be followed, but there is a nuance here too. People who order IT are usually not the stupidest, they do it and can not only listen to what you say, but also see what you do. Customers will very quickly understand that when you promise to do something in a week and every Wednesday on Wednesday, in the middle of the week you send the finished version, it turns out that you need three days to do it. Next time the customer will ask to do it in three days. Therefore, it is important not to expose new expectations.
There is a similar effect of waiting with bonuses. For example, if you pay employees a monthly bonus in the same amount for half a year, they will perceive it as earned. Not as a reward for trying well, but as part of his salary. And if in six months you don’t pay it to him at some point, it won’t be clear to him. He will ask you: “Where is my prize? I'm already used to it. ” Therefore, if you are not ready to raise his salary, you should not pay 10,000 each month, but pay 20 thousand in one month, not pay in another month, and pay 30 thousand in the fourth month. To understand that this is a one-time bonus and it is not permanent.
It may or may not work, depending on its actions, or on the financial performance of the company's department. At the same time, it is very important to rationally pronounce why now there is a premium, and next month there is no. No need just to arrange a lottery. But if you are not ready to make the reward permanent, you should not do it for an average period. And if you always really do it in three days, you can do the task list for tomorrow and report back to the customer periodically not after three days, but after a week. And the remaining two days can be spent on everyone's favorite refactoring and some other useful things.

Source: https://habr.com/ru/post/303002/

More articles:


All Articles