[series_meta]

Silos are eval. There’s a lot of talk about this, but not much action. But just how bad is it? We have identified the silo organization as one the adaptive-anti-patterns. So if you want your organization to become adaptive in this complex fast changing world, you better learn to understand the problems associated with it, and indentify the pattern in your organization.

In this article we explore the effect of local optimization that is aggravated by the silo structure and stands in the way of true adaptiveness. In a follow-up article we’ll explain why the silo structure can bring any growth engine to a grinding halt.

This article is part of a series

[series_post_list_box]

What can you expect from this article?
  • An in-depth understanding of the history and heritage of the silo organization
  • A list of issues caused or aggravated by silos
  • An in-depth understanding of the effects of local optimization from a Systems Thinking perspective

The history of today’s organization

Look at the picture below. If I would ask you what it is, I bet you would say it is an organization chart. In fact, I bet that if I would ask 100 people, they would all say the same. Even more, if I would ask 100 people to draw an organization chart, they would probably draw something like this. Although we use an organization chart to depict the structure of the organization, they all look remarkably the same. Most are a bit wider and taller, but they all share the same kind of structure: a vertical hierarchy, and a structure of vertical silos made out of departments. Every person has his place somewhere in the hierarchy and in one of the silos. Although theoretically an organization could have any kind of form and shape, most people will intuitively draw a structure of hierarchy and silos. The hierarchical silo organization is so prevalent it has become a synonym of an organization chart.
A typical organizational chart

An overwhelming majority of organizations have this form that has its root in the time of Fredrick Taylor, Henri Fayol, and Max Weber. It was in fact Weber who has popularized the term we generally use for this organizational form: the bureaucracy. Weber described the bureaucracy in an essay from 1922 and it still stands today. Look at some of his characteristics of a bureaucracy: 

A hierarchy of authority

A hierarchy of managers control all units of work. Communications and decisions follow this line of authority.

Division of labor and task specialization

People are organized into units according to their specialty.

Management by rules

Rules ensure everyone knows what is expected from him.

These characteristics still describe most of today’s organizations. Despite more recent efforts in decentralization, empowerment of teams and individuals, servant leadership, and the like, the fundamental bureaucratic structure is still present in those organizations. We argue that this organizational structure is not the best way to deal with today’s complexity and fast pace of change. But in this article we focus on a particular aspect of the bureaucracy: The silos.

The silo structure

Division of labor and task specialization still is the dominant factor for organizational design. Sales people sit with salespeople, customer service people sit with customer service people, IT staff sits with IT staff, etc. Today’s organizations are highly compartmentalized into functional silos: sales departments, marketing, operations, customer service, IT, etc. What’s typical about the bureaucratic organization is that decisions flow downward, while information needed to make decisions flows upwards. But there is more to it: the flow of decisions and information also tends to stay within the borders of the silos. Silos have their own management hierarchy, their own KPIs, their own processes and procedures, and even their own language and culture.
Silos lead to myopia

This leads to an interesting phenomenon we have witnessed several times: even if all the individual silo’s KPIs are more or less in the green, the organization as a whole can still be in serious trouble. The organization faces the risk of not identifying the problems because everyone sits happily in their own silo bubble, and in their local world the grass is green. So clearly the problem must be with someone or somewhere else.

Silos lead to myopia

Examples of this can be traced back to the financial crisis of 2008. UBS, a Swiss bank with a 150 year lagacy almost collapsed in the wake of the financial crisis because in the years prior to that their silo structure inhibited anyone to oversay the whole picture and assess risks posed to the entire organization because of the decisions of a part. At the time of the crash, UBS’s Investment Bank chairman, Marcel Rohner made the spectacular admission that the group had unwittingly accumulated $50 billion worth of US subprime mortgage securities on its balance sheet, apparently without the knowledge of the UBS. What followed was a spectacular fall from grace by an institution once considered one the most fiscally responsible in the world. It would have collapsed but for a Swiss Government bail-out.  “Almost anywhere you looked, banks, insurance companies, and asset managers had failed to spot the risks building up in separate desks and departments, because different silos of gigantic institutions did not communicate with each other and nobody at the top could see the entire picture”.2
Silos lead to a culture of blaming

Which is another typical characteristic of the silo structure: it is always another department’s fault. We sales people know how to sell. If only the operations people would fulfil orders in time.

I remember a project I did for a health insurance company a long time ago. Legislation was about to change meaning that all health insurances in The Netherlands were going to be privatized at a certain fixed date. So all health insurance companies had the same fixed deadline to get their products and systems in order to sell health insurances. I was working for a IT services company and we were helping the insurance company with the front-end website needed to process quotes and sell insurances. I will never forget this particular meeting where a colleague and I were present and almost all the other participants were employees representing departments that had a stake in the project. The project manager, trying to get a feel of the current status, directed a question about the status of a certain sub part to one participant only to get a decisive ‘that is not our responsibility’ as an answer. A bit confused the project manager asked the next participant at the large round table. ‘Nope. Sorry. Not our responsibility either.’ With growing amazement we witnessed the question going round the entire table, and sure enough, no one was responsible. No one felt responsible either, except for the desperate project manager, who happened to be an external consultant too. At the time I was bewildered. In a few months from then the entire market of health insurances would be redivided again. There was literally no time to lose, and yet no one seemed to feel just a tiny bit responsible for anything that was not strictly their task.

Since then I have witnessed this phenomenon many times. I am less amazed now because I started realizing it is not the people’s fault: it is the system that is to blame. The silo system ensures everybody performs their own local task as good and efficient as possible, but hardly anybody oversees the whole, let alone feels responsible for it.

Silos lead to a culture of blaming

In theory, a higher level manager should oversee things, according to the bureaucracy rules. But this does not happen because of the silo structure. Most processes that deliver value to the customer involve activities from multiple departments in multiple silos. But nobody manages the value stream. Upper-level managers only mange their silo’s part. We break the customer value driven process into distinctive parts, assign the parts to different departments, build high walls between them, hold everybody accountable for their respective part, and then we wonder why things don’t run smoothly.

Silos lead to longer lead times

The silo structure is also one of the main causes of delays in our organization. We tend to focus on the processing time of the individual activities, but forget about the delays between them. Silos increase the delaus, and therefore lead times, because each silo optimizes their part in the value flow at the expense of others. For a full understand read the Blind for delays article.

Blind for delays

Many organizations are blind to delays. We are so focused on the time we actually spend working on activities that we ignore the waiting time between them. This is a huge problem because usually the waiting time exceeds the working time by far. As a result the delivery of value to our customers is a lot slower than potentially needed.

Local optimization

The above is what Systems Thinking calls local optimization. It is one the most dangerous side-effects of the silo structure. Every division, department, and team gets their own targets. Those targets measure the performance of the activities performed by the department. So, for the sales department we use targets like number of sales closed, or a monetized amount of sales. For the customer service department we mights measure average lead time of customer complaints, number of complaints solved. This will cause the departments to organize their work in such a way that maximizes the chances of success measured according to their local KPI. The idea is that if all parts maximize their performance, the whole must operate at maximum performance as a result.
Interconnected parts

Only, it doesn’t work like this. A complex system is made of interconnecting parts. A change somewhere in the system can lead to unintended consequences elsewhere in the system. Especially complex systems are very sensitive towards local changes. These cause-and-effect events do not stop at the borders of a department. Our silo structure is just an artificial breakup into smaller parts that crosses the way value creation flows. The activities of the different departments involved are still connected no matter how high the wall we build between them.

In a complex system all parts are interconnected

Let’s take a simple example of a company selling widgets. Customers look at the website to find information. The website is optimized by the marketing department to make sure the company ranks high in Google searches. The IT department is responsible for making the changes proposed by marketing. A customer might call the company for more information and speak to a sales support person that provides extra information. A sales rep call the customer back a few days later to see if the customer is still interested. He is and decides to buy. The sales rep takes the order and files it in the sales support system. The operations department then processes the order, collects the widgets from the warehouse, packs them in a box, and labels the box. The logistics department then makes sure the box is delivered to the customer. A few days later the customer calls back claiming he received the wrong products. A customer service rep notes down the problem and promises he will forward it to operations to get the problem solved.

A value stream contains all the activities and people form concept or order to cash

In this example multiple departments were involved. There was a clear stream of activities that eventually led to delivery of value to the customer. This stream of activities is called a value stream: a chain of activities and people that leads to value creation towards customers. It is this entire value stream that determines the perception of the customer towards the company. The customer does not break up the process into separate parts, processed by separate departments. He or she just wants to be supported in his buying process. For him the company is just one whole. This is an important realization: we tend to break up the organization into vertical silos but value creation towards customers is a horizontal process.

Value flows horizontally in spite of vertical silos

Designed for local optimization

These are just simple examples but the problem is that cause-and-effect relationships can be very hard to identify, especially in a complex system. And the silo structure does not help: because we have broken the value stream into separate parts with their own local targets, it gets very hard to see and understand the whole. If we are held accountable for our local activity only, we tend to look inward only. We try to optimize our process: as long as we meet our targets we must be doing fine. We don’t look at the potential consequences for other departments because we are not told to do so. Again, it is not the people who are to blame. It is the way the system is organized. If one department causes trouble for another department, we tend to blame them for it. But bear in mind two things: First, it might be easy to see the consequences clearly in hindsight. But predicting potential negative consequences can be pretty hard. Second it seems perverse to blame someone for not looking at the consequences for others, whereas at the same time we are measuring, reviewing, and rewarding that person only for meeting local targets.

The operation was successful but the patient died

Do you know the saying “the operation was successful, but the patient died”? This very applies to the dangers of the silo structure in complex organizations. The silo structure as part of the bureaucracy has always had a darker side. It led to a more informal use of the term where we blame the bureaucracy for not being able to move things forward. It has become a synonym for the red tape that leads to so much frustration. Another characteristic Weber used for describing the bureaucracy was that it is purposely impersonal, meaning all employees should be treated equally and not be judged on individual differences. We all know how that turned out. But there was some logic in the use of this structure: as long as the work is relatively stable and predictable it does make sense to use task specialization as a basis for designing the organization. You analyze the process, break it into smaller parts, and optimize the hell out of the individual parts. It works because the outcome of the process is more or less stable over time. But our processes are no longer stable and predictable. We live in a complex world and in complex systems the silo structure is a curse.

Towards a solution

There is a very simple first step you can take towards solving the local optimization problem: Give people a KPI one level higher than their local team or department. This way people can only meet their targets by cooperating with others that will have the same target.

System Thinkers emphasize that in a complex system you must optimize the whole to prevent local optimization. But what is the whole? If the whole is the entire organization, we have another problem: that is way to large to expect people to be accountable for and oversee. If we define the whole to small, we run into the local optimization problem again because of unindented consequences elsewhere. In the adaptive organization we take special care to defining the whole to prevent local optimization. We tend to organize around the borders of the value stream or a value proposition.

From silos to value streams

In this article we’ll go into more detail in ways to solve the problems associated with the silo structure in organizations.

Conclusion

The silo structure is a heritage from the bureaucracy design of organizations from the days of Frederick Tayler, Henri Fayol, and Max Weber. And while it once was a justified structure in the days of simple stable and predictable processes, it no longer serves the need of organizations to respond quickly to changes in demand, customer preferences, and competition.

Some issues caused by silos are a blindness for delays causing longer lead times, a blaming culture, and the inability to overview the whole with all its risks associated. This makes the silo structure a serious adaptive anti-pattern.

Many of the problems mentioned are caused by local optimization, a serious side-effect of the silo organization. Optimization at the level of local departments cause a suboptimal flow of value towards customers. Recognizing the pattern in your organization is an important first step. Overcoming the pattern is the topic of other articles.

Bibliography

List of notes and sources we reference from.

Notes
  1. https://www.linkedin.com/pulse/cost-organisational-silos-dr-jacqueline-conway/
  2. Tett, G. 2015. “The Silo Effect: Why putting everything in its place isn’t such a bright idea”. Little Brown, London.
Follow this article

You can choose to follow this article in which case you will be notified when there there are updates of the article. You can also choose to follow all articles of this category. You will then be notified if a new article is added in this category.

Click here to see all the content you are following.