Mickey Granot, Global Managing partner, Next-Era-Consulting
Recently I had a discussion with a group of business leaders that were exploring ways to improve the performance of their companies. I had given a presentation and following the presentation we had a talk. The discussion took off from the content of the presentation but took a “life of its own” leading what seemed to be valuable learnings to all.
The question that triggered this “deviation” was “Mickey, what is your perspective about the evolution of managerial decision-making processes for management of companies?”. Interesting question I thought, I was not sure how it came about, but I had some ideas how I can answer and still stick to the core of my presentation, or so I thought.
“Well, historically organizations exist since the beginning of times. Businesses exist for ages, and obviously all type of other organizations. Since ever, organizations had a goal, and since ever they needed to make decisions towards achieving their goal. And human nature guarantees that whatever we get, will never be enough. So, in addition organizations were always seeking to get more of whatever their goal was, and thus always engaged in searching for ways to become better, to improve. Therefore, since the dawn of times organizations needed to make decisions in two critical areas: delivering to the existing commitments and becoming better. Initially, when organizations are very small, they intuitively make good decisions, as a holistic view of the system is obvious. Then, when organizations successfully grow, more and more entities such as employees, suppliers, customers, subcontractors, investors etc., are joining the system and the holistic view of the system becomes more and more challenging until it totally disappears. However, the fact we can no longer see the system as a whole does not mean we do not need to make decisions which are good for the system as a whole. And the natural way to do that, was to break the large, obscure, and difficult to manage system into smaller, clearer and easier to manage sub-systems. Once these sub-systems are clear, we can push the decision making into these systems demanding the each one of the will optimize its performance in the two critical areas: delivering to the existing commitments and becoming better. It is true it is more manageable this way, but it has one major deficiency; it assumes that if all sub-systems perform as best as they can on both areas, than the system as a whole is performing the best as it can on both. Over and above that, some decisions could have not been made holistically, even if we wanted too, as they either demanded too much time such as deciding how much inventory to hold in each stock keeping location, or too complicated to perform such as exploding the bill of material in production.
It is not difficult to understand that these compromises are exactly that, compromises and as such not only do not lead to the system as a whole meeting its commitments in the best possible way while becoming better, but that it can easily negatively affect and the system as a whole. And I do believe, people, at least intuitively, know that. But were willing to accept these compromises as there was practically no other option.
As a result, decision making processes, tools and practices have evolved to be mostly focused on the sub-system performances and making it easier to make the decisions. And until today the managerial decision-making is based on these fundamentals. Look at things like; product pricing that assumes that the price of product is the sum of accumulated costs in producing it, resource efficiencies that assumes that if all resources are efficient the system as a whole is efficient, product grouping in supply chain and many more such examples. Did I answer your question?” I asked.
“Yes, you did.” Before he had a chance to say anything else, one of the other participants jumped in and asked “This is all very nice, but the reasons that have led to this evolution, are still here today. Even more than in the past; the organizations are larger, have more entities participating, the geographical spread is larger, the governments involvement grew as well. So, even if it is a compromise, it is still the best way to make decisions. Isn’t it?”
“I very much understand why it looks that way, and I think we need to approach answering this question in two separate parts; the first relating to the common approach of breaking the system to parts, and the second to the complexity of the decisions. Let us start with the first. Historically, the evolution of the decision-making process was done based on the assumption that the performance of the system, equals the sum of the performances of its sub-systems. As I said earlier, I do believe people, at least intuitively knew it is not the case. But we human beings we have another trait, which mostly works well for us, still at times not so well. It is the natural tendency of laziness. Our brain is designed to save energy and thus the minute we identify a solution, we tend to stop searching and stick to it.
Even historically, when it became clear that the holistic view if the system is lost, breaking the system down to sub-systems was not the only way. There was one hidden assumption that no one challenged, the assumption that it is impossible to create a holistic view for complex systems. What if this assumption is wrong to start with? What if it is possible to create such a view? And what if it is even easier to operate with such a view versus the sub-system approach? And it is possible.
If we accept that the system performance is not equal to the sum of performances of its sub-systems but rather is a result of their synchronization. It becomes easy to understand that not too many of the sub-systems govern the effectiveness of this synchronization. Which will indicate that focusing on these few sub-systems in both delivering to commitments as well as improvement will guarantee that the whole system enjoys the benefits. It also much easier than the common multi-sub-system management approach as we only need to focus on a few. This was always true, and unfortunately, we stopped searching after bumping on the first solution, and to make things worse then came cost accounting and just rooted these practices into the decision-making systems making the alternative almost impossible to see, or adopt.
As a summary for this part, it is much easier than we think to obtain a holistic view of complex systems, in fact the more complex the system is the easier it is to achieve that. The human tendency of inertia coupled with cost accounting makes the alternative until these days almost impossible.”
I was ready to move on, when I got this question from one of the participants “Why do you blame cost accounting for that?”
“Are you a cost accountant?” I asked, I could see them smile. “Cost accounting had a very important job – allowing the external world to evaluate the performance of the company in the (recent) past. The practices, tools and procedures of cost accounting were developed for that purpose. Not, for decision-making. When you freeze a system in the past, you can do whatever manipulations on the data of the system at that time. Even manipulations that have no ability to really exist in the present, such as what is the cost of providing a service to a specific customer. You can only do that about the past, and even then, some of it will be a guess. One of the things you can do when you look at the past is to answer the question – what was the total cost of operating the system at that time? And now you can take this answer and divide it with any parameter you would like to, for example, the number of units you produced. The result will be a number you would call – the cost of producing a unit. This is only relevant for the past. The big issue now, is that management take this artificial number, from the past and make present decisions about pricing, discounts, deals and product viability. Cost accounting, as it measures the past can easily prove that the system performance is the sum of sub-system performances. It is only true because of the “freeze in the past” and not because it is true. Management today is so based on cost accounting considerations, that I do see cost accounting as responsible for the fact that organizations are not only under-performing, but that their performance is totally random. And before you ask, yes there is a much saner alternative. We should leave cost accounting to do its job – provide the external world with evaluation of the company’s performance in the past and adopt alternative for decision-making. By the way, the alternative becomes pretty obvious if you accept the previous conclusion about the holistic view of the system as a whole.”
“Can you tell us something about this alternative?”
“I can. Briefly.” I said, “If we understand and agree that we have to look at the system as a whole, then it stands to reason that only system measures are important. There are a few key system measures; naturally we would like to look at sales but the truth is that the money that any company makes is not the sales, but rather sales minus what the company is paying to the external world for making these sales. So, the first system measure is the money we make equaling to the money collected from sales to the external world minus the money paid to the external world for making these sales. This measure is called Throughput. All the rest of the company’s expenses are the next global measure. Called operating expenses. Cost accounting treats this cost as variable, which is true but not as cost accounting treats it. It is not varying by the unit produced/ serviced but more in a step function. So, for some changes in quantity this cost will stay fixed. And lastly, we look at the total investment in the company. Now it is easy to make decisions, as all that is required is to ask what the effect of the decision on Throughput, operating expenses and investment will be. Knowing which few sub-systems govern the performance of the system as a whole makes this evaluation easy.”
“Let me now relate to the second aspect of the decision making, the complexity. In the past, in production companies, for example, there was a very large production-planning department. It was needed as customer tend to order fully ready finished goods and are not that interested in ordering based on the components that are used to produce it. Nor do they care about the production process, and lead times. Accordingly, this department would convert customer orders (and/or forecasted orders) to purchasing orders for raw materials, components, sub-contracting, and packages as well as to a production plan. This was a tedious mathematical procedure that needed to be done in cycles until converging into a doable plan. The mathematics were not complicated, the procedure very. So, one compromise was to make all these calculations only once a month. As a result, the minimum supply time to a customer was on month. Even when the company’s production can be completed in a day, or a week. But, to enable such a performance the production planning department would need to be so big, making its cost totally unacceptable.
This challenge had a solution the day the first MRP system appeared. As now, all these complex calculations are done by the computer, that does not make mistakes, can perform the same calculations as many times as needed, and can do it rapidly. You would expect that as a result, the performance to the market will improve. But mostly this did not happen. You know why? Because companies, after huge investments in hardware and software, ran the MRP system once a month. There are companies, that even today still run it once a month.
Similar to that, there are many other complex decisions we used to make compromises on. Basically, because there was no technology that could have done this rapidly, repeatedly, accurately. The thing is that today we have extremely capable technologies which are circumcised by coding into them the decision-making rules that were invented to cope with the fact that such systems did not exist. What do you expect the result to be?”
“Can you give us some examples?” I was asked. “Of course.” I answered, “Take for example product grouping in supply chain management. In the past a company with a large number of SKU’s and stock keeping location could have not handle the multitude of decisions needed if they had to be made for each SKU individually. So, we grouped them reducing the number of decisions to the number of groups. Today, computer systems have no problem in making the decisions for each SKU at each location individually, but we still group them – in the computer system.
In the past to decide on purchasing orders, a team of people worked on collecting and analyzing the data, making an order had a cost. To control this cost, orders were accumulated until the overall value of the order justified the cost. As a result, companies ordered infrequently, and held large inventories. Now, there is virtually no cost to making an order, but the systems continue to accumulate orders to justify the non-existing cost.
There are many other such examples. The sad thing is that we have the technology that eliminated all (or at least most) of the reasons to compromise on our decision-making, but we implement these systems while coding into them the old decision-making rules. This in effect, guarantees that the value gained out of implementing these systems is miniscule.”
“As we are running out of time let me summarize the key points: Compromises are, or at least seem to be, at time unavoidable. But we need to remember that compromises always mean that we also compromise on our goals. Managerial decision-making is sadly based on two such compromises; the compromise on the holistic view through viewing systems via their sub-systems, and the compromise on the quality of the decisions through sticking to decision-making rules that were invented to cope with the fact that we did not have technologies that can make all the necessary calculations accurately and cost-effectively.
Both compromises are not needed, the first was never needed as it is not too difficult to obtain and operate in light of a holistic view, and the second is no longer needed as the computer systems of today can easily provide the needed abilities. And shifting from compromising on your goal, to not makes all the difference.”
