Transcription
Book #107. System Thinking for Managers. Author: Dennis Sherwood. I want to say right away that you will not understand the book by just listening to it, as the video will have a lot of diagrams, on which the book itself is built. The book is complex and fundamentally different from all others, so if you want to understand it, you need to sit down and direct all your attention to the video. Let's go. System thinking is a broader view of a problem. Otherwise, it may happen that by solving one problem, we will get several more serious ones. By mastering this type of thinking, you will be able to make more effective and wise decisions. To make it clearer, imagine that you are flipping a coin. What happens next? Correct, it will fall to the ground. Because only you, the coin, and the ground are involved in this event. But what will happen if you lower the price of your product? A lot can happen, because your customers are connected to the price, and perhaps after this they will want to buy your product and your sales volume will increase, or vice versa, they will be upset because the price reduction deprives them of the feeling of exclusivity. Competitors are also connected to the price, so a price war can begin. You can also add employees who will be given a raise for completing the quarterly plan, or conversely, they will face dismissal, as the company will go bankrupt due to the low price. And of course, the government – which plays a role in regulating trade. That is, a price reduction can engage several units of the system, in connection with which there will be a ripple effect, i.e., one event leads to another, then to another, and so on. With a systematic approach, the entire system should be considered as a single whole. Without dividing it into separate units. Thus, interconnectedness is preserved. With the help of a causal loop diagram, which allows describing complex systems in the form of a chain of cause-and-effect relationships, you will be able to avoid making unsuccessful decisions that seem to eliminate problems but can lead to opposite results in the long term. You will also learn how a complex system will develop over time. That is, you will be able to track the probable consequences of your decisions before you implement them. Thus, you will learn to look into the future.
Chapter 1. System Perspective. We often want to divide a system into separate units and study them separately from each other. But in this case, the connections between the parts will be destroyed, and the working system will turn into non-working subsystems. Therefore, if you want to understand, influence, and even control a system, you must strive for a holistic understanding of it. For example, let's imagine that we are managing a department that is part of a complex system. And decisions made absolutely correctly for your department may not be optimal for the organization as a whole. For instance, you can increase the salaries of key employees to retain them. And maybe everything is fine, but it may happen that another team working on an important project will start to envy them and work not as well as before, in connection with which the project will be failed. Or due to a salary increase, the company may cut costs somewhere, resulting in a shortage of goods and a more significant problem will arise. Therefore, always consider the entire system as a whole, and we will see how this is done further. Systems also have characteristics that are absent when studying individual parts. These are emergence and self-organization. Let's take a team, a team consists of individual parts, team members. The performance of the team cannot be predicted based on knowledge of the performance of its individual members. High-performance teamwork is when all team members function effectively with each other, i.e., when all team members work as one whole. Thus, emergence is when the whole becomes greater than the sum of its parts. And self-organization is when the system itself finds a stable dynamic state. Let's take a flock of birds flying in a wedge. It is unknown how this complex system is maintained. Perhaps the leader informs the flock members what to do, or perhaps such a system forms naturally. In any case, the system finds a stable dynamic state. Even if some bird breaks the formation, after some time everything will stabilize. Or take a cyclist and a bicycle. They cannot communicate with each other, but they self-organize, as a result of which, when moving, they maintain an upright position and do not fall to the ground. A self-organizing system is always ordered and is maintained for long periods of time. A flock of birds is an ordered group, not a random crowd. Or your heartbeat is also an ordered, self-organizing system. At the same time, all self-organizing systems exchange energy with their environment: birds react to air currents created by the movements of their neighbors. And the cyclist injects kinetic energy into the bicycle, which is why they move. Such systems are called "open" systems. Therefore, if you want to create a system that maintains some degree of order and does not disintegrate into individual elements, it must be open and constantly supplied with energy. If the cyclist stops pedaling, he will fall. To supply energy to a department, leaders are hired into an organization. The flow of information within a system is called feedback. For example, football players need to constantly receive and process information about the position of opponents and teammates, so that if necessary, they can pass to the player who requests it. Feedback is part of self-correction. If a cyclist and bicycle hit a pothole, the wheel starts to wobble, but the system quickly stabilizes because it is self-correcting. This effect is achieved through feedback: the person feels that the bicycle is tilting, after which they straighten it. Now let's move on to examples.
Chapter 2. Causal Loop Diagram. Let's build a causal loop diagram. It reflects the main concept – the ability to cope, related to two others – quality of service and error rate. Each arrow indicates cause-and-effect relationships: i.e., the ability to cope affects both the quality of service and the error rate. P means a positive relationship, O – a negative one. From the diagram, it follows that if the ability to cope increases, then the quality of service as a whole also increases, and the error rate decreases. And vice versa, if the ability to cope decreases, then the quality of service also decreases, but the error rate increases. Perhaps you got confused and didn't understand where to put P and O. If an increase in the cause leads to an increase in the effect, then the relationship is of type P, otherwise it is O. The workload of a department depends on the volume and variety of operations that it must process. That is, with an increase in the volume and variety of operations, the workload also increases, and the ability to cope with all this decreases. When errors occur, staff and managers have to deal with their consequences. As a result, employees in higher positions have to work not only for themselves but also for their subordinates, which creates a stressful situation and leads to an increase in workload. The ability to cope is also influenced by the number of effective staff and their training. Computer systems increase the ability to cope by reducing the workload. That is, the more effective the systems, the less the workload, as computer systems allow automatic processing of large volumes of operations. But this relationship has a time delay, because an action rarely leads to an immediate result. There is a delay almost everywhere, but only that which is of particular importance is indicated on the diagram. Staff work, their training, and computer systems cost money, so we can introduce costs. Now our diagram is almost ready. As we can see from the diagram, there is a vicious circle. Which, if you do not eliminate, the department's life will continue to become more complicated. To mitigate this effect, you can introduce a more efficient computer system or increase staff productivity by hiring well-trained specialists. But again, all this costs money. And we need to choose either optimization or cost reduction. Directors usually push for cost reduction, so they cut staff, minimize training, etc. This happens because they do not see that the error rate is directly related to costs. Errors are not free: 1. Costs are needed to correct errors. 2. In a large company, for example, an investment bank – this can cost a lot of money, as you will have to pay compensation to the client if, due to an error, they do not have time to buy shares at a reduced price, and it significantly increases. Therefore, it is imperative to consider the cost of errors, because by depriving the department of trained employees and computer systems, you do not reduce costs, but rather increase them, as the error rate grows. It is necessary to find the level of investment in people, training, and systems at which an acceptable cost level of errors can be maintained. Looking at the diagram, we see the entire system as a whole. But in many organizations, there are a lot of managers who are only responsible for their department and therefore they make more and more mistakes, whatever decisions they make. Each manager is interested in fulfilling their own plan; they do not consider the big picture. Therefore, department heads should gather and discuss issues and problems concerning the organization as a whole. When we build such a system, connecting all individual units into one whole, we will see that if we take action here, sooner or later something will surface there.
Chapter 3. Creativity, Quality, and Cost Reduction. In every business, the question has been asked at least once: "How can costs be reduced without harming the business?" For example, in a television company, if you reduce broadcasting costs, their quality will suffer, fewer viewers will watch this channel, which will interest fewer advertisers, and as a result, the television company will receive less revenue. But that's not all, due to the poor quality of broadcasts, the main staff, namely the stars, will be dissatisfied, which will cause them to move to other television companies. As a result, we get as many as two vicious circles. What can we do to prevent a vicious circle from getting out of control? We can reduce overhead costs and save on the IT department and accounting by outsourcing these functions. Then agree on a quality level below which we will never fall. We can also introduce new broadcast formats that do not involve stars and have a different cost structure, and then find new advertisers or revise current contracts. We can also turn the most successful broadcasts into brands and earn money from selling merchandise, but at the same time, the initial costs will increase, as funds will be needed for merchandise production. Staff dissatisfaction can be reduced by involving them in problem-solving and reviewing compensation. Each of these elements stops the vicious circle and restores control over the business.
Chapter 4. Feedback Loops. Feedback loops are expressed through the presence of one or more continuous, closed loops, loops without a beginning or end, where everything is connected to everything. We have already seen such a loop in the first diagram. Feedback loops are a component of all causal loop diagrams. If you draw a diagram and there is no loop, it means you have not fully described the system. There are two types of feedback loops. First, reinforcing loops. Let's take the main loop in the television company example. We see that with each turn, the situation worsens. This situation, where pressure on costs increases with each turn of the loop, is known as positive feedback, and the associated causal loop diagram is known as a reinforcing loop. But the effect of a reinforcing loop can be not only negative but also positive; in this case, it will be a virtuous circle, which underlies business growth and success. In a reinforcing loop, there are either no O's, or an even number of them. Second, balancing loops. Pouring coffee into a cup, we keep in mind its planned level, say, half a cup. Pouring coffee, we monitor how the level in the cup rises and estimate the difference between the planned and actual level. With this assessment, our brain controls the physical action, in this case, the position of the hand, and the greater the difference between the planned and actual level, the lower our hand is tilted and the higher the pouring speed. The physical action affects the coffee level in the cup. That is, the faster we pour, the faster the actual level increases. And the higher the actual level, the smaller its difference from the planned level. And as the difference becomes smaller, our physical action weakens, and the coffee level in the cup rises more slowly until the difference reaches zero. The form of feedback in which the system strives for a certain goal is called negative feedback, and the corresponding loops are called balancing loops. In balancing loops, there is either one O, or an odd number of them. That is, here we see how the actual coffee level in the cup strives for the planned level. The graph in this case can be depicted as follows. Free links are elements outside the loops but connected to them. In our case, the links are these elements. Links can be input – i.e., representing plans, goals, or tasks to be performed, or external triggers of the system. They can also be output – in this case, representing the main results of the system's work. Free links define the boundaries of the system of interest. There are situations that sometimes exhibit type P behavior and sometimes type O. For example, let's take the concept of generosity towards staff and staff performance. Initially, with increasing generosity, staff performance also increases, but if generosity becomes excessive, performance begins to fall. That is, sometimes this relationship behaves like P, and sometimes like O. Thus, you can expand the diagram to reflect both cases. In systems thinking, these cases are called fuzzy variables.
Chapter 5. Drivers of Growth and Decline. A vicious circle and a virtuous circle have the same structure. And the behavior of the loop depends on how it is triggered. For example, the diagram we considered earlier will look the same. Everything will depend on how the volume and variety of operations behave. If it increases, then a vicious circle will result. If new IT systems are introduced and the volume decreases, then we will see a virtuous circle. From the diagram, in the television company example, a virtuous circle can also result if, for example, the television company releases a show that turns out to be very popular and its revenues increase significantly. Thus, pressure on revenue decreases, and this reduces pressure on costs, quality will increase, there will be more viewers, staff and advertisers will be satisfied, and therefore revenue will continue to grow. For every business, the growth stimulator will look like this, of course, in a condensed form. The larger the base of satisfied customers, the higher the sales revenue, from which we get profit. Profit provides funds for investments, which can be used for new product development, marketing, sales channel expansion, etc. Increasing the base of satisfied customers can be achieved by increasing the satisfaction level of existing customers or attracting new ones. Diagrams can contain explicit and implicit links. For example, we can only assume such links or indicate them explicitly. Currently, "average customer sales revenue" determines "sales revenue." "Total margin" is the total profit from this "sales revenue." "Investment coefficient" depends on how much the company reinvests profits. And "number of new customers per $1000 investment" reflects the efficiency, for example, of a direct mail company. As we can see, free links clutter the diagram without adding any clarity, so it is better to omit them. Reinforcing loops can connect, as we saw in the television company diagram. Resource competition can also be represented by two interconnected reinforcing loops. So, what do we see. Two countries have their own goals. And to achieve them, they both need to use resources. The higher the need for resources for one country, the higher its resource consumption, the same applies to the other. But these are common resources that are taken from one source, which is limited. The higher the resource consumption of one country, the less will be left for the other or for both. The greater the total resources, the more remaining resources. When there are enough resources and the needs of both countries can be easily met, there are no problems. But eventually, there are fewer and fewer resources, so the fear that the other country will not leave enough resources grows. And each country begins to take some actions: it may start secretly stockpiling so that others do not get them, try to outbid prices for resources to secure a reserve for unforeseen circumstances, etc. That is, the need for resources of each country increases. There can be different ways out of this situation, but the most primitive is that countries quarrel and start a war. The number of competitors will decrease, so all remaining resources will go to someone. If several competitors claim these resources, then their number will also decrease, consequently, resource consumption will decrease, and the fear that some country will not get resources will decrease. And the conflict will subside for a while, but it may subside. Now we see three O-type links, i.e., this is a balancing loop. Thus, the structure represents two connected reinforcing loops interacting with two connected balancing loops, resulting in some stability, even if at the expense of some of the countries. But war is not mandatory. First, controlling bodies or a higher authority can be created to monitor the resource consumption of each party. Second, countries can come to mutually beneficial cooperation. In this case, each party must trust each other. And third, jointly discover new sources of resources.
Chapter 6. Setting Tasks and Choosing Goals. Balancing loops always strive for planned indicators or goals. Sometimes movement towards a goal occurs smoothly, as we saw in the coffee graph, but if there are delays in the system, the movement can be jerky. Surely you have encountered a situation where you adjusted the warm water in the shower and wanted to make it a little hotter. You turn the hot tap, then a little more, and then you scream. This often happens in business. Let's take the following example. The budget size is determined during its compilation and becomes a planned indicator. During the year, we may find deviations of actual costs from the budget. This comparison suggests what actions to take to bring actual results in line with the budget. In this case, the necessary action is to change the price either upwards if the actual price is lower than planned, or downwards if the actual price is higher than planned. This diagram reflects both cases, and P and O type links work automatically in both directions. This can also be prices, staff, and much more. When compiling balancing loops, the description of the action must always include both positive and negative words and phrases. Because two actions are possible, depending on the nature of the deviation.
Chapter 7. How to Draw Causal Loop Diagrams. There are 11 rules for this. The first rule: define the boundaries. Everything depends on our interests. That is, if we are interested in an elephant, we can draw boundaries around one elephant; if we study elephants as a species, our boundaries will be a herd of elephants; if we are interested in the African ecosystem – the boundaries should include the entire ecosystem. In addition, add the necessary external links that define plans, policies, goals, external drivers of the system, and its results. Moreover, if we consider the system more broadly, these links can be included in the diagram more fully. There is no universal answer where the boundaries are. You will begin to feel them when you gain experience in building diagrams. The second rule: start with what is interesting. Everything inside the diagram is connected to everything, so it doesn't matter where you start. But if you start with the parts of the diagram that interest you, it will be easier for you to expand it. Ask yourself three questions: What are the external driving forces of the system? What are the main results of the system's work? With these questions, you can determine the input and output free links. And with the third question – what are the main elements related to the problem we want to solve? You will understand which elements should be present in the diagram. The third rule: ask, "What drives this element?" and "What drives this element?" All elements within the diagram are connected by cause-and-effect relationships. The element located at the tail of the arrow is the driver of the element, and the element located at the tip of the arrow is the driven element. Thus, when drawing a diagram, you can move in any direction. The fourth rule: do not overload the diagram. Any element can be a driver of many others or have many drivers itself. You don't need to give in to the temptation to dig deeper and deeper. Choose the most important point that has the most powerful impact, otherwise you will get lost yourself. To choose the most optimal option, do this exercise in a small group, where with the help of several people you will come to a certain agreement. When the diagram is understandable to each of the six people – then you have done everything correctly. The fifth rule: use nouns, not verbs. All the diagrams presented earlier follow this rule. For example, in previous diagrams, it was "quality of service," not "to ensure quality of service." "Ability to cope," not "we can cope," etc. This will make your diagrams look understandable. The sixth rule: do not use terms like "growth" or "decline." For this, you have arrows with P and O designations. And as we have already said, something can first grow and then decline. Using one of these words will mean that the situation will always be the same, and as a result, you will confuse yourself. You can use these words in combination, for example, "growth or decline." Therefore, in the coffee diagram, it is indicated – "physical action," as we can both pour and pour out coffee. The seventh rule: indicate the types of P and O relationships as you create the diagram. Do not leave this decision for later. Consider all relationships as you create them. First, if you don't know which type to put, perhaps one of the elements is incorrectly connected to another or poorly expressed. Second, it will be easier for you to understand the structure and supplement it. And you will immediately understand what type of loop you have. The eighth rule: do not stop. Over time, diagrams will become more and more complex, and it will be harder to understand what is happening. But you must not give up. You will throw many diagrams in the trash, but in the end, you will get one that works. The author spent several weeks on his diagrams. The ninth rule: a good diagram should be understandable to everyone. If people do not understand your diagram or see it in their own way, then perhaps you have done something wrong or missed something. In this case, work on two or three diagrams in parallel to reflect reality as perceived by each group of people. As a result, you will get a single diagram containing various elements from all diagrams combined. The tenth rule: do not fall in love with your diagrams. Do not try to draw a diagram as a clean copy right away, otherwise you will consider it a work of art. Do not be afraid to doodle. The eleventh rule: there are no "finished" diagrams. Surely you will find ways to improve or expand the diagrams from the book, and it will be the same with all your diagrams. Therefore, do not be afraid to change them over time. In fact, this is only half of the book. Now you know all the theory and, in principle, you can create any diagram. In the second half, if you liked the idea itself, you can find many practical examples, thanks to which the information will be assimilated even better. And with that, we will finish, subscribe to the channel, give a like, click on the bell, and leave your opinion about this book in the comments. Bye.