What Should Teams Look Like?

As we undertook this research, we wanted to understand which of the many models of team structure works best for sparking change, getting the work done, and making it sustainable. But as we moved through several rounds of interviews, we came to understand that the field is still in too early a stage to know which model(s) works best. We identified one model that people across the board felt didn’t work, but otherwise saw pros and cons for different organizational structures.

As a result, the question we ended up answering was “With what models are people experimenting?” This felt more true to our research; experimentation is really what people are doing. Case in point: Some teams shifted the models they used over the course of our research. Some are trying multiple models at the same time. Here are different approaches and advantages and disadvantages we observed with each.

Examples: Asheville, New Orleans

In some cities, the IT team has evolved to also be the innovation team. In Asheville, digital services director Eric Jackson has been working to change the IT department’s reputation. “We’re not the typical IT department of no,” he told us. “Our primary orientation is, ‘How can we get you what you want?'” That’s helped make other departments feel more comfortable bringing ideas to him, and has led to city-wide projects that benefit multiple departments.

Hiring someone with no support and no grounding in government is an excellent recipe for leading someone to rage quit.

As CTO of New Orleans, getting the email up and running was task number one for Lamar Gardere. “When we would try something new, folks would say, ‘Okay, that's great, but why doesn't email work?’ There wasn't an opportunity to do anything beyond the basics until we mastered the basics.” Mastering the basics won the IT team trust of other agencies across the city, which gave them the opportunity to dig into city issues beyond email. They started with open data and eventually moved to developing human-centered services.

In the case of both Asheville and New Orleans, because the IT team was already positioned as a cross-governmental, cross-agency entity, with hard work and strong leadership, teams parlayed standard IT projects into service design work.

Examples: Bloomberg i-teams, San Jose

Bloomberg i-team grants help individual cities establish a small team focused on solving a specific issue that spans multiple agencies and/or departments. These teams work to “unpeel the onion,” as Mayor Sandy Stimpson of Mobile put it, on issues including reducing blight, improving economic outcomes for people returning from jail, or reducing the murder rate. i-team members have reorganized entire city departments, redesigned forms, stickers and notices, rewritten call center scripts, used free tools to capture data or coordinate agencies, and in one case altered the state constitution. Whatever a team deems is necessary to solve a problem, they figure out how to get done.

These teams are highly focused. By chipping away at a single problems for years, they often make significant progress. On the down side, this model imposes innovation onto government agencies externally. Because these teams are not invited in––instead they show up at the behest of the mayor––they often face tension with people who have been working in an agency for years or decades. As discussed above, these teams must then deliberately focus on building partnerships and mutual respect to minimize this friction.

A variation of this model is San Jose’s approach. The city’s innovation team met with agency department heads to develop a list of key projects. After factoring in risk, impact, and the level of effort required for implementation, the team came up with a prioritized set of cross-agency projects to tackle as they got up and running. These projects spanned city departments and agencies, but took into account the top priorities across the city.

Examples: United States Digital Service HQ, 18F, Boston New Urban Mechanics, Austin Innovation Office, Orlando Digital Platforms and Service Design

In the consulting model, innovation teams live as separate entities, taking on city-wide projects, or projects within multiple agencies. These teams are called in to help when things go awry, or (best case scenario!) at the start of projects when agencies would like help and resources.

Boston’s New Urban Mechanics team started as an incubator for new ideas in 2010. Over the years they’ve grown to assist multiple departments with specific problems. Austin uses a similar model, working with agencies as varied as the convention center and the recycling center. The team also runs a city-wide certification program on content strategy and user research. The value of this model is that work is centralized in one location. Because the same team works on innovation efforts across the city, they can apply what they’ve learned or built for one agency to the next project with a different agency. Lauren Lockwood explained:

“The needs that our Parks Department might have for reserving a tennis court might be very similar to the needs our property management team has for reserving a room. So it's really important to have a central clearinghouse for needs. Otherwise, you end up with asking your constituents to download a hundred different apps for different city services, when two services might actually be very similar.”

And, because teams are invited in, rather than imposed from the top, agencies tend to be more receptive to ideas. (Or at the very least, less hostile.)

But a consulting model often means a constant hunt for work. While the convention center originally approached Austin to take on a redesign project, other projects have required more wooing and cajoling from the innovation team. In one case, the team decided that in lieu of a holiday party they would do a day of service design at the Austin Animal Center, the largest no-kill animal center in the country. (Party on, Austin.)

The team toured the facility, interviewed staff from the animal center, then designed prototypes and shared their work with the center staff. Based on that work, the animal center began scoping a project with the innovation team. This story exemplifies the lengths teams often need to go to win over skeptical agencies. In our own work with USDS, we too saw that it could take months or years of laying the groundwork with a given agency in order to finally be invited in.

A variation of this model is being applied in Orlando. The innovation team runs a three-day academy that takes front-line city employees through a rapid prototyping and testing workshop for forms specific to their agency. (“We call it an academy because everyone wants to graduate from something,” says Matt Broffman, the city’s director of innovation.) Staff members pair up with people from other departments to walk through their service—for example someone from the Parks Department might try to report a pothole. At the end of the three days, the team launches a user-tested beta site. The innovation team is systematically working through all 350 city services. So far, 70 people have gone through the academy

While this model means a steady work stream, some departments have been resistant to the academy. “They say, “I don’t know anything about computers. Why would you have me do this?,” Broffman explains. But typically the most resistant people end up seeing the biggest benefit. “It’s really empowering to see them do this. To give staff the ability to actually make decisions about how their services are designed and how they operate isn’t something that happens very often in government.”

Examples: CA Child Welfare Digital Service, USDS agency teams, New York City Mayor’s Office of Economic Opportunity

A model we encountered less frequently at city and state levels is embedding an innovation team within a specific agency. This is the model currently being applied by the California Child Welfare Digital Service. Because their work is so new (the agency has been in existence for a little over a year) we are not able to report out any solid findings yet.

A more established team that has applied this model is New York City’s Mayor’s Office of Economic Opportunity. The team has been in existence for four years, and their location within the Economic Opportunity office means their initial work has been trained solely on the products owned by that office, all of which are aligned with the office’s mission of reducing poverty in New York. The team sits side by side with others in the agency—policymakers, technical developers, and data scientists—where they serve as the product and design arm of the agency. This means they they’re an integral part of how the agency functions, rather than an add-on entity who is either sticking their face into problems and annoying everyone, or the team the mayor sent in when everyone else has screwed up.

While there are clear positives to being embedded within an agency, one of the drawbacks to this approach is that the work is fairly siloed. So while the Economic Opportunity team may develop a tool that could be useful to multiple other departments in the city, it can be difficult for those departments to learn about or apply the tool. Additionally, other departments within New York have taken on their own individual innovation initiatives. Economic Opportunity sometimes partners with these teams on projects, but there is no single approach to innovation, product development, or citizen needs across the city, which means that for the end-user, city interactions can vary widely.

One of the drawbacks to this approach is that the work is fairly siloed.

Recognizing this, the Economic Opportunity team has taken some interesting steps to try to spread their tools, knowledge and methodologies to other departments, which we’ll discuss below.

Example: Gainesville

Gainesville City Manager Anthony Lyons told us that the mayor of Gainesville became interested in innovation when he heard about another city’s Blue Ribbon report at a conference. When he returned to town he asked Lyons to look into creating a similar report for the town. The mayor’s primary driver was to make the city more competitive. The city is home to the University of Florida, a school with over 50,000 students, most of whom leave after graduation. Was there a way to stop the brain drain and make Gainesville a more attractive place for starting a business?

As Lyons began thinking about the best way to tackle the problem, he took a look at what other cities were doing, but couldn’t find anyone “doing anything special.” So he turned to the private sector to see how companies were handling innovation. Ultimately he met up with IDEO, and the city brought the company in to help rethink the design of the entire city government from the citizen-perspective. IDEO trained city officials in user-centered design and prototyping, and got everyone from building inspectors to the fire chief thinking about how they could better meet citizens’ needs. Lyons told us:

“I think the difference with what we're doing is not trying to do bolt-on kind of programs or ideas, but to change the philosophy for the entirety of the government and then to adjust the government around that, rather than having the innovation adjust to the government.”

Gainesville’s approach is unique. There is no innovation team. Instead, the city trained people across city agencies on citizen-centered techniques, and all agencies think about their services from the perspective of how they can best serve Gainesville residents. One positive outcome from this approach is that front-line civil servants have begun looking at their jobs differently—they feel empowered to bring suggestions to the table—and we heard from multiple people that the attitude change has been palpable.

While Gainesville’s approach is the most dramatic attempt to create internally-motivated change we encountered, New York City’s Service Design Studio has attempted something similar, on a smaller scale, with its Service Design Studio.

The design and product team housed within the Mayor’s Office of Economic opportunity were pleased with the work they’d done within that office, but wanted to figure out a way to serve other parts of the city. They’d run design studio workshops with other agencies. They had also developed a toolkit on human-centered design that they made available across the city. But they were looking for something lightweight to do every week to support people’s efforts to incorporate user research, prototyping, agile development and other standard innovation elements. Ultimately the team hit on the idea of offering office hours for anyone in any city agency to walk through the Service Design Studio door and get help.

Front-line civil servants have begun looking at their jobs differently—they feel empowered to bring suggestions to the table.

The office hour idea was a shot in the dark—the team had no idea if anyone would sign up, or what kind of requests people might come in with. “We had no idea what we were getting into,” Mari Nakano, the office’s acting design director told us. But they quickly found themselves overwhelmed with bookings. As of early August 2018, they’ve held 103 office hours with 34 unique agencies, and a total of 248 total attendees. The office hours have been so successful that they’ve been getting meeting requests from people outside New York City government—to date they’ve held office hours with 17 other governments including Rhode Island, California, Denmark, and Thailand. The team is still relatively new, so they don’t yet have success metrics to show whether their involvement has positively influenced projects, but they are tracking repeat visitors, projects, and how questions evolve. In the eight months they’ve been holding hours, they’ve seen questions shift from learning about our tools and tactics to discussing specific projects or for assistance in achieving stakeholder buy in.

We observed the office hours on a few occasions, and watched the team meet to discuss topics including an agile RFP, how to bring users into the development of the city’s fair housing policy, and how to get more New Yorkers to report animal bites. We observed, and the team corroborated, that the people who come in generally have some understanding of service design and the steps involved, but need guidance or hand holding on specifics. How do you find people to speak to in doing user research? What’s a quick way to develop a prototype? How do you start thinking about changing the way a process works?

The team has enabled others across the city government to start asking these questions, to shift their thinking around how they offer services or develop policy. But because the team is housed in a separate agency, their reach is limited. They can offer help and support, but cannot guide projects to their conclusion. Whether exposure to the service design team serves as a catalyst for other agencies to bring in their own design and product teams remains to be seen.

Sometimes governments dip their toe into innovation by hiring a few individual fellows, or spending a lot of money to bring someone from the private sector in to launch a team. In general this approach has not worked out well. In some cases it’s left both the agency and the person they hired feeling frustrated and angry.

In the early days of the Presidential Innovation Fellowship program, the fellows struggled to get a toe hold in organizations that often didn’t know what to do with them. Vivian Graubard, a founding member of USDS, said this about the start of the PIF program:

“The first thing we learned after the PIF program launched was that it does not work to send one person in to an agency or a large organization. There's a very limited impact that they can have, acting as lone wolf. It's so much better to have a team where people have various skill sets than having one person.”

Garren Givens, a member of the first class of PIFs and later the head of the program, concurs, and also notes that while there was a lot of energy and excitement around getting private sector expertise into government, after a while the fellows and the program management came to see that making real change would require a different approach:

“There's no way that a fellowship with 100 people in it could somehow tip the balance of where a two or three million person federal service was headed. It became clear that even if we were a 5,000-person technology service, that you couldn't build all the technology that was required to affect change. At some point, you're going to need to work through partnerships in a sort of federated model.”

Sending a lone wolf in also presents issues for sustainability. Aneesh Chopra, the first CTO of the United States, notes:

“You could establish a program where you do a six-month project, but then its lingering effects, that culture and that reinforcement is missing unless you take the thing and sustain it.”

As we’ve discussed, the work can be slow, and it can be a long wait to see one’s efforts come to fruition. So individuals who come in for short bursts to make change often find their efforts are for naught, with no one to sustain the work after they leave.

In addition to the federal level, bringing in one fancy person from the private sector has also presented issues for cities large and small. As Orlando’s Matt Broffman told us:

“For the first year and a half I failed. I was doing a lot of things but producing very little value. That’s not innovation, that’s just work. If you’re stuck in a large organization as one person and not given an explicit mandate, you’re not going to succeed.”

In a few high profile cases, governments heralded the hiring of someone from the private sector only to have them depart quietly a short time later. We heard that the external hirees found it too hard to make change in government, didn’t get the support they needed, didn’t understand how governmental levers worked, or simply found the bureaucracy and speed of government excruciating. In other words, hiring someone with no support and no grounding in government is an excellent recipe for leading someone to rage quit.