When Top-Down Change Isn’t Working, Hand the Keys to Frontline Staff

Kyle Peters is a World Bank veteran with 35 years of development experience. Currently, as the senior vice president for operations, he supports the bank’s goals of ending poverty and promoting shared prosperity, globally. His top priorities: delivering the bank’s lending and knowledge program to client countries, driving improvements in the bank’s delivery model, and finding ways for the bank to be more effective in achieving its development goals. Previously, he held a variety of regional and global roles with the bank.

The breadth and depth of Peters’ experience with the World Bank give him particular insight into the way the organization works. He spoke with The Boston Consulting Group’s Sharon Marcil about the intersection of internal and external goals for an organization like the World Bank. In particular, he describes how he discovered that frontline staff who were passionately committed to the bank’s mission were simultaneously profoundly dissatisfied with their workplace and that the best way to fix the problems rested with the staff themselves. Peters and Marcil talked about the bank’s use of agile techniques to improve the workplace so as to boost employees’ morale, commitment, and results.

Peters’ insights are especially timely, as he spoke to us just weeks before his retirement from the World Bank. BCG has had a long relationship with the bank and with Peters, and we feel fortunate to have heard his thoughts at this milestone. The World Bank’s agile program is now being managed by the CEO's office, and a second group of agile volunteers is being recruited. The team is working to roll out a handful of changes across the organization by the end of 2017.

You’ve been at the bank for a long time and you’ve seen a lot of change efforts. What situation prompted the most recent one?

Change has been difficult at the World Bank. We are considered to be one of the world’s premier development institutions. We have 90% client satisfaction and a lot of support from our shareholders. It can be hard to launch a change initiative in an environment like that when, from a lot of perspectives, things are going well. Around the same time, we faced the need to be more cost-conscious and implemented a new matrix structure, which made it harder for people to collaborate, particularly across sectors.

When we implemented our new global matrix structure, we realized that we didn’t have a sophisticated way of laying out how we were working internally and what our issues were. We knew everything about our clients, with extensive datasets and scorecards, but when it came to our own work, we were relying too much on anecdotes. So, we started by taking a good look at ourselves. We examined everything—our operational processes, workload, spans of control, performance management, culture, collaboration, and knowledge management.

What did that diagnostic reveal? Did it include a particular catalyst for turning to agile as a solution?

A few things came across loud and clear.

First, we learned that our staff loved the “what” of the World Bank. They loved the mission and the impact that we had on clients. But they hated the way they worked. Our staff survey showed 80% to 90% satisfaction on mission and objectives but only 25% satisfaction on internal processes and the way we make decisions. Those results signaled a downturn in people’s faith and trust in management and in their satisfaction with the way that we worked. Everyone was frustrated and felt powerless to get real change started.

Second, there was a clearly asymmetric risk versus reward culture. This is common in the public sector. We don’t have bonuses, we have difficulty exiting staff, and we have a healthy fear of reputational risk.

Third, we have a sort of academic culture. Our way of working is more academic than business oriented. A lot of people have PhDs, and they are used to going into their offices, closing the door, and working on the perfect product.

We started doing some analysis. For instance, we asked BCG to look at the span of control of our managers. They had to string together five internal data systems to get an answer. We figured out that frontline managers manage an average of 30 people, 50% of whom were five to seven time zones away. We are a knowledge organization, but we had spans of control for managers that were equivalent to those of call centers. We also looked at workload analysis, and we found that we didn’t have enough transparency into how we managed workload. A lot of work was concentrated among a small number of task team leaders or project managers. We did an analysis of meetings and found inefficiencies and a lack of attendance discipline.

We had to change something. The way we’d managed in the past was unsustainable. We decided there had to be a better way, and we got a lot of input from a lot of sources to help us figure out the best path forward. And that’s what got us to agile.

Why were you attracted to the concept of agile and the bottom-up approach to it?

We had spent years trying to do simplification from the top—I would change something, and people would say “Oh, that’s good,” but when we really implemented the change, we found in many cases either that it didn’t work or that other changes complicated things further and overwhelmed any positive impact. The second issue was that no one working on any of the changes had actually prepared a project in the past decade. People at the top didn’t appreciate what the frontline staff experience was really like.

So I decided, let’s let the staff guide changes from their vantage point. I took a bit of an institutional risk. I got in front of the staff and I said “Okay, I’m going to give you the keys to the car. You know how to drive and you know the rules of the road. You know that we have certain rules and policies that we have to follow. But I’m going to get out of your way and let you just start these experiments. Start changing things, and let’s see what you come up with.”

People didn’t believe me at first. There were many meetings where they looked at me as if I was crazy. Eventually, though, I signed a blanket waiver that waived all the bank’s processes and procedures. Basically, I was saying to my staff: you tell me how you want to work.

So, that’s how we got to this bottom-up as opposed to top-down approach.

We understand that you’re ten months into the journey. How has it gone?

We immediately saw firsthand how this could work. As an example: When we appraise loans for our clients, we produce documents that go to our board for approval. The documents that we were sending to the board for each project were about 150 pages. For the entire time I’ve been in operational policy we’ve been telling people that we want shorter documents. Nothing seemed to work. But when we turned it over to the staff piloting agile, within three months they produced a 40-page document that they were happy with. So were our board and the client.

There were several other examples, all showing the power of allowing smart people who know how to work and know the rules to just start working. I think for the staff, it has really been empowering. For me and others in management, it’s been really gratifying to see people start doing it.

How are you going about the process of taking success in one part of the organization and scaling it horizontally across the rest?

Scaling has been one of my sources of frustration. For instance, we had a peer review

process that was developed in agile that became quite useful to people in the pilots. But when we tried to scale it horizontally, people had a negative reaction because they didn’t understand what we were doing. They hadn’t been on the journey that had resulted in this particular innovation, so they didn’t see the value.

I realized that if I had a group of people who were working together and decided to do something in a new way, that’s great. But if you take the way they’ve decided to do it and then tell the next group, “You should do it the way the previous group did”—well, you’re back to the top-down problem.

There has to be some socialization around what’s being done and why. One good thing I saw happening is that the people get excited about the changes they are making and then everyone wants to participate.

This plays into the other frustration of doing this type of thing in the public sector. It takes a lot of time. So, you tend to get very impatient and you want to push, but then you realize that it is about people changing the way they work, changing the way they think, changing the way they work together. That takes a lot of time, particularly in an institution that has worked in a certain way for 70 years with a great deal of success. You have to let each group solve real problems and find the right solutions. But as we continue, we’ll learn to communicate, communicate, communicate about what we’ve learned, and our successes will make other groups less skeptical and more open.

What about this change effort are you most proud of?

We decided early on that we needed consultants to help us, but I knew that consultants couldn’t be the face of this initiative. So, I wrote a blog for our intranet that read: “Help wanted.” Basically, we were asking for ten World Bank staff to take a year out of their career to help us launch this agile initiative. We ended up getting more than a hundred people who were willing to step outside the normal process and come help, which was great. I think there was a lot of excitement in the ranks, which I was really proud of.

Before we started agile, staff at the bank were very frustrated with the way they were working. And they felt reluctant to make suggestions and escalate problems. Opening the door to transparent conversations about where things are stuck is beginning to make a real difference.

Another result of our project—which has an upside and a downside—is that now you can’t go to a meeting in the World Bank without a senior manager saying the word agile. Groups that didn’t want to attempt change are now knocking on the door to participate. So, something has really caught on!