Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage Interviews Systems Thinking for Management

Systems Thinking for Management


1. I am Shane Hastie from InfoQ and I am privileged to be here with Esther Derby. Esther, I know you fairly well and some of our audience will know you, would you mind introducing yourself for the rest of us?

Sure. I started my professional career as a programmer, I haven’t made my living writing code for some time, because I was so good at writing code that they made me a manager, and shocking as it may be the skills that make you a good programmer do not necessarily match the skills needed to be a good manager. So I had to shift my focus and really learn a lot about people, about systems, about organizations, and I’ve focused my attention for the last couple of decades on management and organizations and teams. So I am a member of the board of the Agile Alliance, we are here at the Agile Alliance Conference and I am delighted to be here talking with you.


2. Thank you very much. You have been focusing a lot on that manger’s role and more and more likely the manager’s role in self-organizing teams. Can you tell us a little bit about that?

Sure. I think early on in the life of Agile methods there were some people who said "once you have Agile methods you don’t need no stinking managers". I beg to differ; I think managers still have a very important role to play in organizations and even when you have self-organizing team. So I have been helping managers make the transition from a more traditional traditional manager-led team to how their work changes when they are focusing on creating the conditions for self-organizing teams and how they interact with those teams.


4. So the manager creates the environment?

Well, at least sets the conditions right. Then the team can work within them to shape their own environment but the conditions need to be there. So setting creating the conditions where a team can form and self-organize is a very important thing, developing people, I think that’s a critical role for middle managers and working across the organization to make sure that the organizational systems are working. Instead of just looking down at what my individual team needs and looking up to see what my manager needs they need to be looking across the organization looking to the right and left of them and understanding how work flows through the organization and how to create and adjust those systems so the work flows smoothly to the team and the policies and procedures and structures in the company enable the team to self-organize. So looking across the organization, developing individuals, and creating the conditions for teams.


5. What are the traps? In one of the sessions that you are actually delivering here, are team traps. Can you give us a small flavor of that?

Well, there is a set of team traps which I will talk about in a minute but there is also a set of traps that managers fall into; and part of it is a trap that is set by the organizational structure. If you look at hierarchical organizations there are particular patterns that can play out for people at the top of the organization, people in the middle of the organization and people at the bottom of the organization. And one of the traps that middle managers get into is that they are pulled, they are pulled by the demands of the people above them that they report to and they are pulled by the demands of the people that they support. So in that pulling they often feel very isolated and because they are competing for scarce resources in the organization, whether those are material resources or financial resources, or promotions, or their boss’s attention, they often fall into competition with each other.

Instead of looking across the organization to make it possible for everyone to be more successful, they look at how I can be successful. To a great extent this is driven by the organization; it’s easy to look at it as a character flaw but it’s really not related to their character, it’s a behavior that is driven by hierarchical systems. So that’s a trap that is very very easy for managers to fall into. If you are interested in learning more about that, some of that theory comes out behind of the work of Barry Oshrey (???) so that’s a trap.

Another trap that managers fall into is that if you are management material, if you have been promoted, then to show any sign that you don’t have the expertise, that you are lacking knowledge, that you need help, that you are confused, is seen as weakness. And so, middle managers are very often reluctant to ask for any of those things; further their own education ask for help, band together with other middle managers because that could be seen as a sign that they are not management material. So that’s a big trap for managers, very hard to admit vulnerability. And that’s true in many, many organizations particularly US organizations.

So those are two traps that they fall into and then they fall into a trap of not recognizing that if they want to achieve the benefit of the team effect, of a self-organizing team effect that they really have to create real teams. And there is an old adage that calling a tail a leg does not make it so. They are no five-legged dogs because calling the tail a leg does not make it a five-legged dog; so many organizations pressure managers to have teams where the membership is constantly changing, or teams where it’s not clear who the members are, so it’s very easy to just call something a team when it doesn’t really have any of the characteristics of a team and isn’t going to achieve the results that a team can really achieve.


6. So what is a real team?

Well real teams have a hand-ful of characteristics: they are usually small in number, usually less than ten people because if you get to be more than ten people it tends to break down into subgroups because the communication overhead is just too high so people simplify it by breaking into subgroups. Small in size; they have a common and compelling work goal that takes all of their skills and all of their efforts to achieve it; they have complementary skills, sometimes slight overlap but complementary, so that they are working in an additive way, not just they are all doing the same thing and integrating at the end. Compelling all and they are mutually helpful. So a handful of characteristics.

And then they need the supporting elements, so that they can be successful. They need to have adequate material support, so in terms of physical space, equipment, tools (high tech and low tech), they need to have appropriate education, for the areas where they need additional skills they need to have access to outside expertise, and they need to have a connection back to the organization. And I think that that’s something that often gets overlooked when we look at self-organizing teams, it is that they are there to serve an organizational purpose, and if there is not a customer who is eagerly awaiting their output I have to wonder why they are there. But they also need feedback back from those customers to make sure they are making the right thing, getting acclamation and know where they fit in the big picture of the organization.


7. One of the things that I’ve heard you talking about and mentioning lately is this thing called "Seeing and steering systems", and you’re talking about that at the Agile Conference.

Yes, I believe that people always choose from the best option that they see is available to them, given their current state of mind, and under the relentless pressure of day to day events in the organizations, people very often just respond to what’s coming at them, you know "The server is down", "Someone’s leaving", "we have to hire", so they are bombarded by these events. And when you are responding to events it is very very difficult to see the patterns. So my session about seeing and steering systems is about stepping back so that you can see the pattern of events, not just the isolated events. And once you’ve stepped back you can begin to understand something about the underlying structures that are really driving behaviors in the organizations. And that’s where your leverage is as a manager.

It’s not in telling people what to do, or telling people to work harder, or work smarter, or get more done with less, it’s really an understanding of underlying structures and the patterns and shifting those to enable everyone to do their best work.


8. How does the manager who’s under the constant pressure of somebody’s leaving, server’s down, the immediacy problems, how do they step back? What’s some practical things people could do?

If something needs to be responded to, if the server is down you need to look at why that’s down. They looked for patterns and one place managers can see patterns is if they are using a Scrum of Scrums to coordinate the work of many teams, they can look for the patterns that come up in the Scrum of Scrums. So that’s one place they may find patterns. They may find patterns coming out of retrospectives or they may just start jotting down notes about what seems to happen frequently and what’s present when it occurs.

So, it doesn’t have to be a big elaborate thing, it can be looking for the data we have and seeing what patterns are in it. It can be jotting down notes and say I think I see a pattern, they have a hunch, let me go think about what the factors that might be involved are; but it does require taking a little time to breathe, a little step back so I think it’s important for managers to schedule some time for reflection, to schedule time where you can sit and think.

I have a very good friend, actually a mutual friend Johanna Rothman who said that she became a much better manager after the birth of her first child because she was breast feeding and so she had to sit by herself in a room and pump every day, and she found that while she was doing that boring task that her mind was working but she wasn’t interrupted so she had that built-in reflection time; but don’t wait to have a baby to do it, to start it. Just start it, find yourself twenty minutes a day and just jot down some things you notice about what happened and soon you will begin to see patterns, and you can investigate further. So it doesn’t have to be a big massive data gathering just a start.


9. Is there a risk with looking at patterns and seeing the wrong patterns? I am thinking about Blink, Malcolm Gladwell’s book ((

Yes, that’s a very interesting book. Well I think there is perhaps another trap in that often people ascribe to individuals what are actually system problems, and ascribe to systems what are actually individual problems. So it’s important to have some data to back up your hunches and to consider both possibilities. Because very often what looks like an individual problem is actually related to a systemic issue. Another colleague of mine was telling me about a situation where he was working with a number of teams in an organization and they noticed that, a manger noticed that the philosophy for one particular team was quite volatile and her hunch was that it was this one guy and he wasn’t doing a very good job at checking in his code and so on and so forth.

So she thought that if she got rid of him things might get better, but she took a step back and realized that all of her five or six teams in her area that they all were volatile in that same pattern. So that led her to see "Oh, it’s probably not this one guy" and it turned out it had to do with the way the backlog was being groomed and structured and so that pattern existed across the organization. So rather than firing this guy or telling him "work harder, work better" she was able to say "Oh, ok the problem is over here and I need to work with it in a different way in a different part of the organization". That’s very often the case that what shows up in the individual performance is actually related to something else somewhere in the organization.


10. You recently have been writing what they don’t teach in management training; I suspect that some of this is covered here.

Well let me tell you about the management training I went to. I was a programmer on Wednesday and a manager on Thursday and they sent me to management training where I was instructed on how to fill out a staff requisition form and who received the pink copy and who received the goldenrod copy and who received the blue copy. Of course we have advanced now, somewhat we no longer have those quadrate forms but often people are taught the checklist tests rather than how to think about them.


11. They have a webpage now.

Right, they have a web page for that now, but that doesn’t help in terms of how do you think about hiring somebody? How do you think about what the skills are needed? And so forth. So some management training is really poor like the one I received so I had to educate myself, but I find it there are very few management training courses that are really focused on system thinking and understanding how to create an environment where people can be successful, and how to reason about what’s going on.

So that means that managers are left with their only lever is really to tell people to change their individual behavior performance and they end up in a position where the feedback about how the system is working is delayed and when you have delayed feedback you may pull the right lever but you may pull it too hard which then sends the system into an oscillating pattern. So the whole notion of how to reason about your system I think is sadly missing in most management education that I have seen. Maybe there are some out there that I don’t know about that are fab, I would be glad to hear about them.


12. Thank you. Moving back to the team stuff, you talked about management traps, can we talk a little bit about what are the team traps? What are the traps that these high performance self-organizing wonderful Agile teams can fall into.

Assuming they had the conditions for success are there: they are a real team, they have a compelling purpose and they have the support they need, just getting the ideas that working in a group of humans there are things that people fall into.

They have to do with process churn where people either don’t have enough ideas or they are churning about arguing over something they don’t have the requisite variety to come up with a good solution or have too many ideas or having ideas at the wrong time. So that is one trap that I see a lot of teams fall into, particularly teams in our area where we have a lot of smart people who are really good at generating ideas. They are under a lot of pressure in which case they might alight on the first plausable idea that comes out. So that whole notion of churning round ideas is something that happens to teams.

Failure to navigate conflict is a huge trap for teams, and I think many people equate conflict and confrontation and they really are two very different things. And paradoxically conflict has the potential to really build trust. I think people worry that if we are going to get into conflict it’s going to destroy trust but the ability to navigate trust congruently actually builds trust. That’s another trap. Withholding information is a trap that teams fall into and I am not talking about information about where the code packets are or how to program in Java, but information about how our working relationship is going. I talked to a team where everybody on the team was very concerned because one of their coworkers had very strong body odor, but instead of telling him they hinted, they said "Hygiene is important" and they left a can of air spray on his desk, and he didn’t know what was going on and they kept getting more and more annoyed because they felt like "We are trying to tell him but he doesn’t get it".

They finally got to the point where there was a division, between the teams. So, they were withholding some important information, and when someone learnt how to tell him, he was embarrassed, but grateful that he knew and that he could do something about it. So withholding that kind of information about relationships,. Withholding information about your internal state. I worked with a team where one of the guys on the team had been really steady ready, reliable kind of a guy and then suddenly he took another style and he started withdrawing and nobody really knew what was going on.

The information he was keeping from the team was that he had something really very stressful going on in his personal life and he was embarrassed about people finding out and he didn’t want to be a burden and so he was withholding that information but it was keeping him from being active in the team and keeping the team from supporting him and helping him. So, once he was able to find a way to talk about it, then he was able to be present and people were able to take up a little of his slack because they knew once this was resolved he would be back.

So that was a big thing, but very often it’s people withholding information "I am confused about the discussion we are having," so someone gets left out of the discussion gets behind or "I disagree with that" but I am not saying it, so I am not really buying into what the team is agreeing to do." Or "I feel that my idea was just dismissed so I am going to stop giving my ideas". So there is all sorts of information that is withheld that is critical for the team to function well.


13. That’s a scary level of honesty. How do you create a safe environment?

It can be. Well, you have agreements, and I think that having working agreements help people understand specific protocols, and that addresses kind of the working together level. There’s another idea called simple rules which comes out of the work of human systems dynamics that says we have some rules that will help us guide our behavior across many situations and across many levels of the organization. And one of those might be, we are honest about what’s going on with ourselves, which will drive a certain level of behavior - but make it a known thing that it’s just a normal thing to do. Making conflict (not confrontation) but making conflict a normal part of work, helps create that sort of safety.


14. Can we distinguish; can you explain for us the difference between conflict and confrontation?

Yes, conflict is just when people have any difference in priorities or needs or wants or thoughts. And that’s inevitable that’s just going to happen. Sometimes you only need one person in the room to have a conflict, because people are torn, out two minds. So that’s inevitable and it’s going to happen. Confrontation is the thing that people are really frightened of, where voices are raised and hurtful words are stated and judgments are thrown at other people.


15. How do you tackle conflict without it becoming confrontation?

I think the more we can put it outside ourselves, instead of it’s in us and it’s personal, and put it outside and talk about it, the more it’s likely to be handled in a productive way. And recognizing that there are a handful of sources of conflict then you can start thinking about "Well this seems like this is the source of the conflict, and here are some strategies on how I might handle that". There is relatively few sources of conflict in most organizations; they may be driven by the structure, going back to a kind of the classic example: testers are rated on finding bugs and programmers are rated on getting the software to done at a certain stage and those two things can be in conflict; you don’t see that on Agile teams, but in many in traditional organizations that’s the norm. So that’s driven by structure so that people attribute bad motives to those people.


16. "Testers are evil".

Or "developers don’t care about quality". So that’s a structural conflict very common in organizations and once you realize it’s driven by the structure it’s easier to depersonalize it and see they are not evil they are just working under different set goals.

People taking a position focus, so arguing for one particular solution versus what are many solutions that are possible to solve this problem and choose from them; just misunderstandings about simple words, sometimes I ask people the question "What does one hundred percent of the time mean?" and surprisingly for some people it’s not one hundred percent. So you say we are going to always refactor, and your definition of always isn’t one hundred percent of the time then that’s going to lead to a conflict and it’s just a misunderstanding about words.

So another source of conflict is people having different values and deeply held values, and unlike a position focus where you say "Ok, let’s see how many additional actions we can come up with and what are your interests behind this" people are not going to change their deeply held values. So then we need to look at where can we find some common ground? Where can we find some additional solution that would meet different subsets of values? That’s not quite as common in the work place, that’s not as common in the workplace as some other conflicts.


17. What about intra cultural teams?

Yes, that would come up in those situations and then it is really important to look for - I am thinking here - this is an interesting subset of a conflict because people will observe a behavior and they will attach an interpretation to it and then they will attach an evaluation to it. And the interpretation and evaluation are always from the person’s own cultural lense and so it there is behavior that looks different, then the motives are ascribed. So for example if someone doesn’t bring up concerns in a meeting in a culture where people are used to speaking out forcefully, they may interpret that as - "he didn’t bring up those concerns, he wanted to undercut us". Which is an interpretation and an evaluation. When it may just be that in the culture where this other person came from unless you are the senior person in the room you don’t speak. So, that’s a whole nother subset of issues that come up around culture.

There is some interesting work by a group of people called Cultural Detective that helps people find the bridges between we share some similar values and this is how we express it differently. So that is sort of interesting subset of conflict. But it’s sort of related to, or similar in that it gets ascribed to a person’s conflict about preferences and that shows up in the workplace a lot. Again my friend Johanna who has a very pronounced preference to have things planned, and settled, and I have a very pronounced preference to have things open. So my way of planning a vacation is my husband and I decide what state we are going to and pitch the camping gear in the back of the truck and go.

She has hotel reservations and knows where she is going to be and when I tell her about my vacation plans she starts feeling ill and vice versa. So that sort of difference shows up in the workplace and often leads to conflict. And if any of these conflicts isn’t solved at the level where it starts they all devolve to what gets often labeled as personality conflict where people begin to ascribe angelic motives to themselves and evil motives to the other person on the other side of the conflict. And those can become very entrenched and toxic within a team. Which is why it’s so important that teams have some tools to solve those conflicts productively and navigate that conflict. Because once you get into one of those lock on hate relationships, where I will do anything to make you look bad in a meeting and vice versa then it becomes a spiral and it can tear a team apart.


18. What are some of those tools?

Well you have to come to my session; I talk about that, I think we don’t have time to go into that.


19. We’ve been running out of time. Just one final thing which is: you as part of your sponsorship to the conference you gave away card deck with useful hints inside.

Yes, I like to do a little card deck every year this is the third year that I’ve done a card deck. And this year the card deck is "Agile team setup for success" and it talks about the enabling conditions for teams and the need to be present to reach high performance and I have written a lot about it on my blog ( and it’s all there.

Oct 11, 2011