To learn more about Jeff Sutherland and his company, Scrum Inc. go to www.scruminc.com
I read your blog and there are so many interesting things there. One of the things I noticed was the Happiness Metric – it seems like an unusual metric but it seems to be working very well. Can you tell us something about that?
It is very interesting because we are constantly learning new things to do as part of our Scrum that actually accelerate the Scrum. So I first heard about this in Europe – a company in Sweden I was working with had implemented the happiness metric. They did it because it was a better forward indicator of revenue than the traditional financial metric, which I was looking at the past. Counter-intuitively that metric simply isn’t very good at predicting the future. Here’s why the Happiness Metric is better at doing that: it is based on how people feel about our company and how they feel about their job. And if they are feeling good about it, they are projecting into the future, they are anticipating whether the company is doing well and whether they are going to do well in their jobs. And that is a leading indicator of financial success. The cover of Harvard Business Review this week is about happiness as an index of performance, so this idea is becoming widespread.
After seeing it implemented in Sweden I came back to United States and we were just starting up a new SCRUM team in my company last December. So I said, “Hey, I think we should implement the Happiness Metric.” The way we are going to do it is we are going to rate how we feel about the company on a scale of one to five, five being great, one being terrible and then we are going to rate how we feel about our job on the same scale. Then we talk about what would make us feel better. Scrum has a retrospective at the end of every SPRINT, where you evaluate, look at what’s good, what was not so good and what process improvements should we implement. What I found is that this is a much better way to identify ideas for process improvement in the retrospective.
At the end of the retrospective people were talking about how they feel and what will make them feel better. We were doing weekly Sprints. The first week the top impediment was communications; we were just getting started up. By the second or third week the focus turned to user stories. The team said they would feel better if the user stories were better coming into the Sprint. That was the top thing on their list. So we attacked that first.
I decided to use the pattern – Scrumming the Scrum – testing improvements that were made just like you would test features. Put it into the backlog with acceptance tests. I had seven tests for better user stories.
The results were dramatic. After the first Sprint, the team’s velocity had grown from 50 to about 70 points. They said the stories still weren’t good enough. We put it in the backlog again. 90 points. Still not perfect, said the team, so we went at it again. 130 points. Within three weeks, three Sprints, the team’s velocity had nearly tripled. That happened buy simply by asking people how happy they were.
Here’s a quick graph of my team’s velocity. The graph of weekly Sprint velocity showed an average of 20 points for two people until December 2010. We added two more people and it jumped to fifty points, and then tripled using the Scrumming the Scrum pattern.
So I said, “Wow, we should keep doing this, its working.” So we used the Happiness Metric each and every week. And we used it to Scrum the Scrum each and every Sprint. There’s some oscillation in the graph, but basically the team’s velocity went up about 10% each week. And sure enough, by the end of December we hit 260 points.
So that’s the value of the Happiness Metric, you can take an entire company and multiply its productivity by 500% a year. So imagine if all the companies in the world did that – every department multiplied their productivity by 500% in one year.
Speaking of the happiness metric and in general about SCRUM, it increases engagement so much, but why do you think people are so disengaged in the first place?
If you look at surveys about how people feel about work, most of the time people just aren’t that happy doing what they’re doing. Deloitte did a survey last year – global survey – what they found was 80% of the people that are working day to day in a job, no matter what job they are did, they are not motivated. They feel like they really don’t have any control over what is getting done, they feel like they are not treated as important people. They feel like work is boring and all they want to do is go home at five o’clock.
So that is the fundamental problem. And I see it everywhere. And it’s because we have management that is really not competent. So it’s like a coach of a soccer team. If you had a coach of a soccer team and that soccer team went out in the field and they wouldn’t pass the ball, wouldn’t play hard and kept losing every game, what would you do with that coach? You get rid of that coach. But for some reason we think it is ok for companies to have people that are not motivated, they don’t have a really good feeling about this company, they don’t feel like the company is going anywhere. Yet they don’t do anything about the management. And what’s happened is that it’s added up to a global financial crisis. That’s how bad it’s got.
So everybody on the planet is hurting over this. There was a very interesting article in the Harvard Business Review May 2010 and it was entitled the “Wise Leader” by Takeuchi & Nonaka. And in that article they asked why did we have a financial crisis? They said well, we did because the leaders of our companies used bad judgment. So why did they use bad judgment? And what can we do to fix it? Well we need to train leaders that are able to bring more transparency in the organization, more honesty, more commitment, more teamwork and at the end of the day, Takeuchi says the wise leader implements Scrum. So they perceive Scrum as a solution to the global problem. So Scrum is really bigger than you think. You wanted to talk about the future of project management; it’s really a lot bigger than just Scrum in software development.
You have been a leader for many organizations, you have founded Scrum. What are some of the most important decisions you make as a leader in these organizations that you lead?
The most recent one happened just over a year ago. Until the end of 2010 my company was embedded in OpenView Venture Partners, a venture capital group. And I had more things that I could handle, I was busy constantly. But the CEO of Venture Partners kept saying, “Jeff, you should expand your company Scrum Inc., if you feel the need to bring Scrum to the world, you need to get a team to help do that. You need to do more than just consulting and training and so forth. What other companies are doing is not enough.” So I said, “Ok.” And I expanded the team. So we started using Scrum to run the company. When you start doing Scrum, you have to start giving the team responsibility, you have to help them. Takeuchi said that in every great team, they saw some primary things that would manifest themselves.
One was a sense of autonomy; the team itself is responsible for its own work. They are choosing their work, they are estimating their work, and they are executing their work. And they are not reporting to a manager, they are reporting to the customer or product owner. How can we make the customers really happy? Ideally, when you are first starting Scrum, you have people who want to volunteer to do it. Or, if you are starting a company, don’t hire anyone who wouldn’t volunteer. So we had a few good people who were really fired up. Now as soon as they start working you have to really start to listen to them.
One of the things that Taiichi Ohno, who invented the Toyota production system – he has a great little book on the workplace and how it works. It’s a set of talks that Taiichi gave to project leaders at Toyota. And the first talk, the basic idea of the talk was the project leaders want to know, “Taiichi Ohno, how did you become such a great project leader, you created the Toyota way and you are making Toyota the number one automobile company in the world? What is it that makes your work so great and how can we be great?”
Taiichi Ohono said he started with two principles. First, “One of the first things you have to learn is about half of what you think is wrong.” Now you come into work everyday and you think you know the plan, you think you know what you should do, but actually when you go to do the things, they never work out the way you expect. And it’s because the way you thought about the problem simply wasn’t correct. And so being a great project leader or running a great assembly line you have to constantly be inspecting and adapting. You need to say, “I need to do some experiments to see which half of what I think is wrong.” Now he says this is really important because if you are trying to do continuous improvement and changing things, it’s very uncomfortable for people. You have to have the trust of the people to lead them through the change process. So you have to be open with your experiments and constantly look for what does not work, and when you find it you need to confess to the team you were wrong. He said if you do that with your teams, they will follow you. That’s the first principle of project leadership.
Ohono’s second principle is that you have to recognize is that half of what your team thinks is right. So every day you need to come into work you need to be looking for what are they thinking, what they are saying, what they are doing that is clearly the right thing. When you find that you need to tell them how great they are and he says if you do that they will follow you anywhere. That’s how you become a great project leader.
You mentioned Taiichi Ohno, Takeuchi & Nonaka, are there other people that had a tremendous impact on you as a leader?
Yes, you know there are several different threads coming into what is currently known as Scrum. And one of them comes through micro-enterprise lending. Nobel Laureate Dr. Yunus in Bangladesh was looking at the World Bank donating hundreds of millions of dollars to build dams and bridges and he saw that the people were still starving beneath the dam after it was built. So he saw a lot of time was spent on these huge projects, sometimes the projects would not even be completed. He said we need to get the money to the people. These big projects are not the solution. So he started working with small groups maybe five or six people who were too poor to feed their children. And he coaxed them to come up with a simple plan where maybe they could buy a fruit cart and sell fruit the town, maybe they could buy a sewing machine and start to sew clothes. When every one of the five or six people had a plan, they’d all talk about it together then he would give them the equivalent of $25 as a loan. It was a loan; they had to pay it back.
Then he let them go out and execute their plans. Then they would talk regularly and when everybody on the team paid their loan back he would loan them another $25 to take their business to the next level. What you see is people who did not have enough to feed their children, now had money to feed their family. Once they had food they could buy clothes. That means they can send their children to school. From the small business that they started, all of a sudden they are building a new house. And all this can happen in less than a year. So I was asked to be on the President’s advisory committee of Accion, a non-profit organization in Cambridge Massachusetts that was modeled after the Grameen Bank in Bangladesh. They worked mainly in South America and Central America. So I learned about what they were doing. They wanted me to bring my entrepreneurial background in startup companies to help poor people because they thought that would be really useful.
When I worked with them I realized we were really freeing the people from their chains. Because of their inability to work together and help each other, they had no food to eat. Getting some help and restructuring the nature of the system and a little project management, all of a sudden they could be successful.
And during that period, every day I was working at a software company and saw that the developers never made enough software, their projects were almost always late, and as a result management was unhappy. And it occurred to me these guys are just like poor people who are stuck in a system that doesn’t work, why don’t we try some micro-enterprise lending inside a software company? The result eventually evolved into what we call Scrum today.
So that’s how Dr. Yunus influenced me and Scrum.
But I want to make a quick aside on just how miserable developers were using traditional project management. I met Peter Green in 2005. He was the agile leader at Adobe and he started rolling out Scrum after we met. A few years later he showed me the analysis of the effect, he actually did a paper on it if anyone’s interested. Anyway, his analysis showed the last release at Adobe using the traditional waterfall method. It took a couple of years to get done, and it had thousands and thousands of bugs. Then he showed the first two Scrum releases – done in half the time, less than half the defects. He said you know this is interesting, but they don’t reflect the real personal issues in the company. During our last waterfall release four people died from the pressure. In the two Scrum releases we have had since then nobody died. That is the real difference between Scrum and traditional project management.
You mention Professor Yunus, we were planning to get him as a speaker for a PMI event, that didn’t work out, but its an amazing analogy that big dam projects fail and micro lending succeeds in the same way that traditional project management fails and Scrum succeeds. I had not even thought about it that way, but now as you were speaking it struck me.
Because what we are really asking management to do is – let’s invest just a small amount, like two weeks of work, let’s just structure the team that they can all talk about the work and then we give them permission to be free for those two weeks. Let’s not follow them for two weeks and let’s see what they have at the end of two weeks. But management has to make an investment and give up a little bit in order to allow the people to start the work and bootstrap themselves out of poverty, the poverty of bad software and not enough software.
Let me give you one example. A few years ago Yahoo launched hundreds of Scrum teams. After a couple years the financial people did an analysis, and they were interested only in cost of software. They said we don’t want any soft factors. Not revenue, not happy customers, just how much did it cost to produce software? And because most of the company still used the waterfall method, they wanted to know how much Scrum teams cost vs. waterfall teams
It turned out that the software from the Scrum teams cost 37% less than waterfall teams. With 150 teams, that saved them $200 million per year – free money for them. They were really excited about this, but I thought it was terrible, because it wasn’t good enough. Now there were 50 of their teams that were well coached and they were implementing Scrum well, where they had good engineering practices, their software was done at the end of the Sprint, they eliminated all the roadblocks. Those kinds of teams got 300-400% improvement. The other teams didn’t do all that. They only did some of it, basically pretty bad Scrum implementation. But still, even doing it badly get’s you 37% more. I would say more than half of the teams on the planet that are doing Scrum are in that range. It’s not as good as it could be, but it is saving the world billions of dollars. That’s a lot of what I am doing now, going round the world, helping people, training, consulting, motivating them to actually to take Scrum to the next level. Everyone should be doing 5-10 times better than what they were doing before Scrum. That was the design goal of Scrum and we have studies showing every team can achieve this if they implement it well.
SCRUM has gained so much in popularity and this year even PMI has come out with its Agile certification. Now when you go around the world what kind of resistance do you find now in adopting SCRUM? Are they the same objections that existed some years ago, what are people afraid of in adopting SCRUM today?
I think more and more people are realizing that they need to be agile. They don’t know what it means, but they know they want to go there, they know that they need a better process. You have companies around the world like Microsoft, where 3000 people are doing Scrum and in their instruction manuals for using their tools they say everyone should use Scrum. A lot of people are saying, well they are doing it, but how do we do it. But as soon as you look at it, it is a big change. So we find it is easy to do it at the development level. There are few developers who do not want to do it, but in general Scrum provides a much better life for developers. They have a lot more fun because we want the agile teams to go at a sustainable pace, have a normal work week, so they have a life outside work. So a much better life for developers, that’s usually where it starts.
But the developers can only go so far without the help of management. We are doing more and more work with the management teams. They need to make a lot of changes. They need to start asking the teams, “How we can help you? What are the things that are getting in your way?” If it takes six months to get a purchase order signed, it’s hard to get anything done. That’s the kind of management dysfunction that we have in a lot of companies. And the managers are responsible for fixing that. They have to take responsibility for running their companies the way they have never done before. And that is going to change things.
Then you go down to the middle management, and they ask them what they do and they say well I have a department that I am micro-managing, that’s what I am paid for, to watch over these guys day to day, to tell them what to do, track everything that they are doing. And if I’m not telling them what to do, that we have a product owner to do that, if I’m not checking up on them every day and reading reports and so on, because the teams are managing themselves, what am I supposed to be doing as a middle manager? This is threatening to my job.
I spent over twenty years in the software development world. But I made sure I wasn’t always the one responsible for the team. Instead I coached the teams to be responsible for themselves. As a software developer that had to be a manager I didn’t want to spend 80% of my time in meetings and going around and checking up on people. That is no fun. So I built a system that checks up on itself and what the teams do is they report on their work to themselves and their customers. Then I spent 80% of my time removing the problems that the team has. So that has made me four times more productive as manager. I have 80% of my time to actually fix stuff, instead of 80% of my time going to meetings which are largely a waste of time.
I have been going around the world, asking people, “Think of the meetings you’ve been in the last week. What percent of the time was productive? And what do you think they tell me?
5 – 10 %?
Some startups are focused on getting stuff done, they say about 20%. But in any big company, you go into its 10%. If you to the Department of Defense its zero percent. What we need to do is free the managers from their chains. They are going to all these meetings and wasting their time. As a manager if you are wasting your time, you are not feeling good about yourself. You are not going to feel very good about the company. That’s going to show up in the Happiness Metric which we talked about when we started. But nobody is doing the Happiness Metric. Everybody is busy in those meetings being unhappy and not talking about how to get happy. My advice to these managers is that you can be a great person, a great contributor to your company and to your society if you didn’t have to waste all that time in these boring meetings. Think about the opportunity costs. To change means managers they have to change the way they think about their role and that is very hard, it is really tough.
To learn more about Jeff Sutherland and his company, Scrum Inc. go to www.scruminc.com
If you liked this post, please share it with others and sign up with the Future of Project Management to receive regular updates in future.