Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ


Choose your language

InfoQ Homepage Interviews Pollyanna Pixton on Agile Leadership

Pollyanna Pixton on Agile Leadership


1. This is Amr Elssamadisy at Agile 2009. I'm here with Pollyanna Pixton. Thank you for joining us, Pollyanna! What would you like to talk about today?

I'd like to talk about leadership, especially leading Agile teams, but more importantly what senior leaders do to help support their Agile teams in their organizations. Actually, we are finding the same principles are effective everywhere.


2. When you say senior leadership, do you differentiate it from regular leadership or is there a difference?

There is not really a difference in my mind, but I'm really focusing on is how leaders that are command and control can stay out of the way, step back and let teams and everyone below them on teams at all kinds of levels make their own decisions and take ownership and deliver.


3. Do you have any specific advice you'd like to share with us about that?

There is a lot of issues that you can do. Collaborate is one, but mainly you need empower the teams, let them make the decisions, let them figure out how they are going to do it, stop asking for reviews, stop asking for status reports and don't give them the solutions. The hardest part that we're seeing with people that come up the ranks and become leaders - whatever you want to call those in the organization - or leadership that forms inside the organizations, is that people come up to the ranks or become leaders because they solve problems well. Now there has to be a shift in their minds. You have to stop solving the problems and you have to start asking questions to help people think in a different way to solve the problems, when you do it without giving them any solutions.


4. Is this specific to Agile teams or all teams?

It's all teams, it turns out.


6. What do you see as important?

It's really important what they are finding in some research at Stanford right now. If anybody gets in a position of power over someone else, starts making decisions based on less information. As leadership and leaders move up in the organization, they should be making less and less decisions. You can go into a group of people or team of people and say "How many of you know which processes are broken?", they all know. "Do you know how to fix them?" "Yes, I know how to make them better." Why aren't we letting people do the improvements they need, solve the problems they need, change the process to make them work more efficient? The answers are in your organization. The people in your team know how to fix the things and what to do. Leaders are just to enable them, power them, move the obstacles from them and leave them alone to do their job.


7. There are 2 things in there that you said that really grab my attention. Number one is the use of empower and leave them alone, if you will. Empower is one of those oft used phrases that when I say it or I see other people say it, people saying "Yes, we heard that 100 times!" What's different? Are we asking about the same empowerment that people have been talking about for 20 years and hasn't been working?

It's a little different. I guess it's you have to give ownership as leaders, at the ones that can do it. You give ownership - I use what I call authentic motivation which comes out of the Alfie-Khan work and being punished by rewards He believes in the 3 Cs - 1.Collaboration. That means let people collaborate in making decisions themselves. 2. They need to be able to choose their work. We can't often let people choose what and when, but we can always let them choose how. So often leaders are saying "Do this and here is how I want it done" - that's micromanagement and that is demotivating terribly.

We hire the very best people and we treat them like crap. What are we doing? We hire good people and we have to give them the responsibilities, let them figure out how and let them do it. The final one is giving people content or interesting work. We don't have that problem in the software industry. I've been working with an architecture firm. They don't have that problem either, they have interesting work, but you have to just keep it interesting for them. Those are the 3 Cs - they give people ownership and motivation and then you can stand back and let them do their work. The standing back is hardest thing leaders are struggling with.


8. What do the leaders do if they stand back and don't involve themselves?

They ask a lot of good questions to help the team stay focused, they ask questions without giving solutions, but help them view it in a different way, they create a safety net, so people can fail safely. They remove all the obstacles a team may have that are required, they make sure they have all the tools they need to succeed. But they don't make decisions and they don't solve the problems!


9. Leaders don't make decisions?

No. I built the entire stock exchange. We spent 5,000,000 dollars a month and it failed twice before I got there. 2 and a half years later it was running. We built it and it's been running for 15 years. I made one decision because, I as a leader, was technically dead. I was too old. They were using new stuff that I didn't know about, but I made one decision and that was to let one person go because the team leader came to me and he said "If you don't take Andy off the team, they are going to throw them out the window". That's the only decision I made and in a huge project like that one.


10. One of the leader's main jobs is to ask quality questions to help the team do what?

To help the team make their own decisions and find their own solutions.


11. Why does the leader need to ask the questions?

Because they usually will get stuck or start going down the pathway, it may be the wrong way or they may be frustrated and not be able to find a solution because their thinking may become too focused. Often times somebody will come into a leader's office and say "I don't know how to do this" or "I'm stuck here. What do I do?" Most leaders would say "Have you tried this?". As soon as they say that, they've given them the solution, based on their experience and then, they've taken ownership. Therefore, the person does what the leader gave him the solution instead of finding even maybe a better one or another one. As soon as you ask for status report, you take ownership, as soon as you ask approve or review, you've taken ownership.


12. Ownership is a key factor to successful teams and you lose ownership as soon as you start making decisions as the leader. Therefore, even if you know the answer - let's assume you do - don't ask the question because you'll be taking ownership away, which is key to a highly productive team.

That's exactly right. What you can do is ask - my favorite question when people come in my office, in the old days - "How would you like to solve that?" or "What do you want to do about it?" or " What do you need from me to solve that problem?", but not give them the answer "What can I do for you?" In my workshops, we spend some time talking about all those questions.


13. Linda Rising said something to me last year and it went something like this "We go into a project, we help them succeed and when we are most successful, they have a hard time paying us because they say 'We did it all! What do you want to get paid for?'" I hear a lot of similarity. Is that really what a successful leader is, where you help the team do everything themselves?

That's exactly right. What's a leader's point? What can we offer? Technology we start dying, we can play, we can lead up, we can keep the distractions away from the team, we can create a culture that fosters trust, we can remove the fear from the organization, we can remove all the obstacles, we can make sure - like I said before - everyone has what they need to succeed, we can create a place where people want to be, not have to be. When we do those things, with integrity and authenticity, then we are great leaders. But we don't make decisions.


14. If you don't mind, tell us about building culture and trust.

Culture and trust. Well, you can't force anybody to trust each other - that's a done deal. But there are things leaders can do to create a culture of trust. The first one is when you are working with teams, if there is broken trust on the team, you have to take care of that, because if you keep a person with broken trust on a team, you have to make the decision whether you want to help them build trust again - broken trust is like a rope that has been broken - you have to put all those little pieces back together. It's a lot of hard and long work. Often times it doesn't work. I think it was Marcus Buckman in his book First break all the rules - I think - that talked about the fact that suspicion is a permanent condition. Once you distrust someone, it's very difficult to get it back.


15. Much more difficult than new persons build trust.

Yes. Usually you have to take care of that. You find out if the team needs them, if they don't need them, you have to put them in another seat on the bus and probably on another bus. If you have a team that needs to build trust and you have to foster trust, the first thing you have to do is to get rid of any of the debilitating fear. In collaboration, according to Warren Bennis, people are fearful of 3 things: losing their identity, losing their intellectual mastery and losing their individualism.

In the software industry, people have a lot of fear about losing their intellectual mastery. As soon as they start, we've seen it with pairing, we've seen it with teams, we've seen it in a lot of places that people say "Just let me work all alone. I'm a pro, I can do it. I don't need to work with anybody." They are afraid to lose their intellectual mastery. As leaders, when we know those, we know how to deal with them. You can acknowledge people, you can reflect their identities, you can do those kind of things, the team can do those things. Then, I ask that the teams start doing a task together, that involves everyone, a simple short win and then they celebrate success.


16. What involves everyone is the key.

Yes, they have to involve everyone. They have to find everybody in the team to work together. Then, measure results and only results, no individual measurements, team based measurements. Israel Gat talks about the idea of letting the team evaluate themselves and then they decide whether they are going to share with leadership or not.


17. Team based measurements help build a culture of trust?

Yes, they do, because people will start working together. Take for example Jack Welch whose probably one of the worst leaders of all time. He operated in a non-trustful environment, because he would every year fire the bottom 10% of his people as low performers. If you are right above that level, you're backstabbing each other to make other people look badly - that's distrustful - so you will be one of the people that don't get fired and the other person next to you will get fired. That's how it operates. It's disincentivizing for collaboration, it's disincentivizing for team performance, so you need team based measurements. In IBM, there is a kind of a funny award that they are giving to several people in the organization - 20,000 dollars. If you run an Agile team and 20,000 dollars hit the desk of the person next to you, how do you feel? "How come I didn't get that?" It's not a good reward, it's not a good recognition, it's not a good measurement.


19. There is another thing that you said - removing obstacles is one of the things that a leader should do. Yet, what I believe I heard you say earlier, is that a leader shouldn't have something like status reports. It's not necessarily the same thing, but how does a leader see obstacles if he/she doesn't know the status of the project?

The team will tell you. If you trust them, you give them ownership, you say "When you need something from me, let me know", they'll do it time and time again. This is eye opener, this is the light bulb for leaders, especially executives, because if you really think about it, if you trust your team, they'll come and tell you. If there is no game, they will tell you the truth. If you have an honest truthful organization, it will work.


20. I'm assuming that you've seen this work again and again.

This is not theory. I've been doing it for years. There is a lot of companies that you can reference now. Semco, Ricardo Semler's company, Whole Foods is doing the same thing, AEG, Johnsonville Foods. Johnsonville Foods are now letting the teams hire and fire their own people. The leaders don't get involved in any way, shape or form. There is the REI - that whole organization.


21. Are those organizations successful organizations? They are making just as much money as the other guys?

Yes, they are. CISCO, the same way it is happening within CISCO. There have been some recent articles written about their CEOs and their leadership CIO leadership and how they chose for collaboration and trust. The most interesting thing for me about building a culture of trust is 1. it's free and 2. the data which came in this year from great place to work institute for the fortune 100 companies is that the top 100 most trusting companies compared with the next 100 most trusting companies - the top 100 outperform the next 100 by 120 % based on 2004 - 2008. Over 4 years they've doubled their performance based on trust and that's one of the key factors. It's kind of like "Why are we not doing this?"


22. Because there has to be a leap of faith? Because trusting is scary?

All of those things could be true, but the payback is huge.


23. What else is really important for effective leadership in teams? You said stepping back.

Stepping back and stepping up. Leaders have what I call the leadership tipping point, that you just have to figure out or sometimes we refer to it as a leadership 2-step, because it's one to step up and one to step back and you have to find the tipping point for yourself. You have to know what I usually ask people to do is figure out what the red flags are for their teams when they know they are in trouble and then to figure out how to help them without giving them the answers. Don't ask them what's wrong, because they say one of 2 things "Nothing's wrong" or they will tell you everything and they will give you a whole bunch of stuff.

Sometimes we do what we call zero gravity thinker in the room - somebody that knows nothing about what they are doing and come in and look stupid so they can ask questions and just ask them to explain what they are working on and what's the ultimate goal, what's going to happen, how they get there. Zero gravity thinkers really help. Leaders can't do that, they can have somebody else do it for them, they can't do it for their own teams. I've done it as a leader for other teams and I love it, but that's all you can do. Then, you have to be very clear about when you try and step up and step back.

We have a tool we call the macro leadership cube, which you define together with your teams. Exactly what they see is the boundaries or the constraints or the ideas mark up windows, purpose whether it's differentiating to gain market share or just to keep market share, but results has to be one of them. Considerations can be one and a whole bunch of things, but results always has to be one of the sides of the cube. Then, as long as the team stays within the cubes and doesn't bounce up against the sides, you leave them alone. It's a real tool to help leaders as soon as they start coming out of the cube and then you step up and help them figure out what they are going to do to stay inside the cube.


24. Two thoughts come to me when you describe this. Number 1, you probably need visibility to know when they get out of the cube and number 2 is previously you said that when teams are in trouble they'll come to you so it almost sounds at odds of you monitoring the teams and as soon as they are getting out of the cube.

I don't monitor them. They know we've decided on the edges of the team together, so they know when they are bouncing up against that and they'll tell you. When they'll know they've crossed some of those boundaries, they'll tell you. That's not the issue. The issue is still you trust them to tell you when they are vibrating against the walls. But the idea is to be very clear on what's going on. We've been using this idea and teaching this idea in IBM to their people and now there is a little mantra that goes around saying "Stay out of my cube!" It helps them with the micromanagers. Leave them alone.


26. So, we are clear about when they come to me?

Yes, they are clear about when they are in need or trouble.


27. What the boundaries are, almost.

Yes. That's like letting them setting up their own red flags, as well, for themselves. It's all about giving it to them - ownership.


28. What else can you tell us about teams and leadership? We just talked about stepping up and stepping back and you step up when they are outside of the cube, when they come to you. Previously you said something about an open environment. Is that different than trust?

Trust is part of an open environment, but the idea is in an open environment you need to create an environment where people are free to push back, free to express their ideas. There is non-judgmental, open and honest communication, safe, fair - those kind of things are created in the environment.


29. Non-judgmental? That sounds almost something that is too good to be true. The stereotype of a corporate environment is very judgmental.

It is, but the Java believers did not let that filter into the team, they still keep it separate.


30. Do you have advice on how to do that?

How do you do non-judgmental? That's a good question! I guess, from a leadership point of view, you ask people questions. "Why do you feel that way?" or "Let's talk about what contributions everybody brings to the table."


31. I've heard you say leadership, leadership, leadership - the noun, not leaders the people. Is leadership a role that we all take on or are there leaders and the reason I ask is I guess you just talk to the people. I'm thinking a CIO is not going to talk to everybody in IT.

They are not, but leadership is all over the organization. Often times, leaders naturally emerge, there is no doubt about it and they evolve, they change - some different places and different pieces. New leadership will arrive and new leaders will evolve and come forth. Leadership is all over the organization. You can do leadership with your pairing programming, you can do leadership with reaching out across the silos, you can do leadership with marketing, you can do it all over.


32. Everything we're talking about here is basically for everybody?

That's right. Not just for these upper types. They do have a tendency, as they go up the organization, to lose the basic principles of leadership.


33. Why is that? You say it "tongue in cheek" - I'm assuming there is a lot of truth behind that. Why is that?

I think it's for their failure. People don't want to lose control, because they are afraid they will fail. Higher up they go in the organization, the more visible they are and the other part is there are fewer and fewer jobs up there. They are very competitive at each other. The idea of working in teams across the organizations are not very good. There is some organization and I can't remember which one it is, it's mentioned in Good to Great, where the leader was so great at fostering collaboration and non-competitive environment with these vice presidents that they didn't have to worry because most of them could go out and get jobs in other places and he was happy with that.

He got them working there and then, when there was no place for them to move up in that organization, they could move in the other organizations and that's fine, but they became great leaders of other companies. We have the idea that if I don't make it in this company, I can't make it anywhere else. A lot of people say to me "I'm not going to be valued for this kind of leadership, for this kind of work in my company." I said "Then, why do it? Because it's the right thing to do." There are a lot of places in our work life and our private life that we can use great leadership.


35. That's powerful. And it will be scary for leaders.

Yes, it will. When I first went into IBM, they said "How are we going to get our leaders to transition to this kind of leadership?" I said "Ask them how they want to do it. Get them together in the room, let them collaborate about what they are going to face, how they want to solve that problem and let them discuss it and they will figure it out." Why should I be telling them how to transition? I don't know.


36. And hence the question at the end - "Why are we paying this? We figured it out ourselves!"

Exactly right. For me, that's a success factor. That's the biggest compliment I can get. That's perfect! But I don't believe in going into organizations and transitioning their leadership. I believe in going into organizations and giving them a set of tools and let them use those tools to help solve the problems at their own organization. My job is to get out of there as soon as possible. I don't want to be in there forever. Please, get me out of here! I don't want to do this! Please let me go home. No, that's codependent and that's bad leadership. It's not it. I tell people "I don't understand your culture, I don't understand your processes, I don't understand everything else, but I can give you some tools to have a more collaborative, more effective, more productive organization. The results are huge. They can use these tools right after they take the class.


37. I know that you have a new book and I'm assuming a lot of what we've been talking about today is in that book. Is that true? Tell us about the book, please.

That's true. The good thing about it is we wrote it collaboratively. That was a very interesting idea. Addison Wesley approached me and I said "Sure, I'd be happy to write a book" and my business partners, Kent McDonald and Todd Little said "You can't write a book on collaboration by yourself." I said "OK", so we collaborated and we wrote it together and it covers how to make better decisions using conversation and collaboration around a business value framework. It talks about collaboration, leading collaboration, collaboration process, stepping up, stepping back, leadership tipping point, build the culture of trust - all the stuff I told you about today. Read the book! That's it.

Jan 09, 2010

Hello stranger!

You need to Register an InfoQ account or or login to post comments. But there's so much more behind being registered.

Get the most out of the InfoQ experience.

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Community comments

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p

Allowed html: a,b,br,blockquote,i,li,pre,u,ul,p