Four cities across the world are preparing to launch the “GovCombinator.”

Jean-Louis Sarbib (left), Dennis Whittle (middle) and Nathaniel Heller (right) are promoting a new process for solving city-level problems. They call it “GovCombinator.” (Christopher Swope)

WASHINGTON, United States—On the fifth floor of a glassy office building here, just a few blocks from the White House, there’s an experiment in workplace serendipity that may some day change how cities solve problems.

In a spacious office suite with rows of desks and widescreen computer monitors, 160 people employed by 32 NGOs and social enterprises go to work. In one way or another, all are engaged on issues related to open government and transparency, either in the United States or overseas.

The space is called the OpenGov Hub. It’s the creation of Nathaniel Heller, who led asmall accountability-focused group now based there. It started as a place for bootstrapping organizations like Heller’s to break free of their tiny offices and enjoy a roomier space with broadband connections, videoconference hookups, and a room for hosting events.

What the initiative turned into is a place where social connections can flourish. The lunchroom became a place for likeminded people to network while heating up pasta. Ideas began to hatch. New friends began introducing each other to potential funders and partners. For a handful of groups based at the Hub, such conversations coalesced into the founding of a new startup focused on creating a more bottom-up approach to development aid.

“First it was an infrastructure win for all the organizations here,” Heller says of OpenGov Hub. “But then we started to see joint projects and joint problem-solving and some of the groups coming together to do things without any engineering.”

Co-working spaces are nothing new, of course, and Heller knew this wasn’t the first time a startup had been born in one. But the Hub’s success as a catalyzer—and the work so many of the NGOs based there do with city governments—got him and one of the startup’s founders, Dennis Whittle, talking one day over lunch. What if you could bring the same methods and serendipitous collisions of ideas into local government? What if you had city officials co-working alongside open-government advocates and technologists? Could it make for more creative public policy?

Policy sprinting

Those questions led Heller and Whittle to a new idea. They call it the “GovCombinator”—the name is a takeoff of Y Combinator, a Silicon Valley seed fund.

The concept is to create a co-working space where high-level city officials would sit side-by-side with people from civil society, business groups, and the other stakeholders needed to solve discrete urban problems. The exact mix of people would change depending on the issue under consideration, which might be something like deciding on a new bike route or figuring out how to implement a new ordinance.

The idea goes beyond office space. The second key component is a deadline. Work at the Combinator would be condensed into a “sprint” of perhaps 100 days. This concept is borrowed partially from the world of software development, with its focus on achieving iterative improvements that build over time. It’s also borrowed from Nadim Matta’s Rapid Results Institute, which has had success with training and facilitating teams to make real progress on problems like homelessness or AIDS prevention over short periods of time.

The GovCombinator is an idea that still exists mostly on whiteboards. But it’s beginning to get its first test half a world away in Nairobi. Meanwhile, Heller, Whittle and other partners are working on launching pilots in Mexico City, Kansas City, and in their own backyard of Washington. The idea is to have the network of four cities share their experiences and, if it works, to grow to more cities.

The idea has gained the attention of Rose Gill Hearn, a principal with theBloomberg Associates consulting group. She has been able to help generate some interest in the project with cities she works with, as well as offering feedback on how it might work.

“What GovCombinator does is it embeds civil-society actors in city hall, so to speak,” Hearn says. “So you have civil society working together with the government officials who have the power to change policy, do a contract, set an initiative, and act on the information coming from very valuable resources.”

Not a task force

It may be easier to describe the GovCombinator idea not by what it is but by what it is not.

It’s not co-working in a traditional sense. Most people working out of the Combinator would only do so for 100 days at a time and probably only part-time. It’s also not an incubator or an accelerator, as the goal is solving policy questions, not launching a startup. And it’s not a retreat—there would simply be too much actual work taking place.

Heller is also careful to note that GovCombinator would be different from a task force. “We’re intentionally trying to not do that here,” he says. “The big difference is task forces are often internal to government. They’re often not empowered to do a whole lot in terms of implementation or execution. We’re trying to expand the circle of influence and bring in a host of nongovernmental actors.”

Another comparison Heller steers around is the “hackathon,” where cities team up with software developers for a couple of days, often a weekend, to build new mobile apps. “What happens the Monday after the hackathon or the policy jam?” Heller says. “Ninety-nine times out of a hundred that’s the end of it—there’s no follow through and a lot of disappointment.”

On the other hand, if GovCombinator is successful, “it can be where ideas from those hackathons and policy jams land on Monday morning,” Heller continues. “Let’s slot it in for the next sprint. We know what the problem is, but give it 50 or 100 days to really sort it out.”

What most separates the GovCombinator concept from these other participatory systems is the intense focus on building social connections. “We’re talking 100 days, plus or minus parts of a day, week after week,” Heller says. “What we’re after here is to build up the soft tissue of trust between people, the social capital. We’re trying to bring personalities and people and their aspirations and fears and wants back into this process of citizen engagement. And our theory is that regular co-location is one way to get there.”

The Kaya experience

While the Combinator is not quite live anywhere yet, some of these ideas have been getting a test run in Nairobi. Al Kags, the co-founder of a “think-and-do tank” called the Open Institute, runs a co-working space in the Kenyan capital called the Kaya. It’s modeled partly on the OpenGov Hub in Washington, with flavors of other tech hubs in Nairobi. (See this Citiscope story from April on Nairobi’s tech sector and accelerator spaces.)

One of the biggest issues in Kenya these days is the implementation of a devolution law passed last year. Authority over many government functions is passing to 47 newly created counties. Kags sees this moment, as county officials work on figuring out their new responsibilities, as a ripe time to test some of the Combinator concepts.

“There’s a lot of pressing problems,” he says, noting that government and civil-society organizations in Kenya have historically had an adversarial relationship. “There’s a need for people in the policy space to come together and work together, and a need for people in government to work closely together with civil society, academia, and the private sector.”

As Heller did with the OpenGov Hub, Kags worked hard to curate the mix of people at the Kaya. There’s a handful of civil-society groups working on devolution questions, as well as consultants, a young political party, and an association of freelance journalists. When county officials from around Kenya are in Nairobi, they’ll often spend a day or two working out of the Kaya.

“We’ve succeeded in enabling the idea of sharing a space with government officials, who come on a regular basis to hang out at the Kaya,” Kags says. “When you have formal meetings, the conversation has to address this or that. Here, we’re catalyzing informal conversations between the various sectors.”

Currently, a group at the Kaya is four months into a six-month sprint focused on building transparency structures into county governance. One product of the sprint is an online portal called the Devolution Hub, a clearinghouse of information related to Kenya’s devolution law. Another product is the Open County Dashboard, which aggregates county-level data and allows for comparisons. There is also a lot of discussion around the best frameworks for transparency.

One lesson Kags has taken away from the experiment is that it can be tricky to get the social mix right. Initially, a tech startup working on devolution was based at the Kaya. But he says there was a bit of a culture clash between the younger and sometimes loud tech workers and the civil society and government people, most of whom were older. The tech workers have moved out, Kags says, noting: “We agreed to keep the Kaya as a space for policy wonks.”

Kags is also finding that it’s a lot of work managing the space and facilitating the social dynamics within the group. Initially he thought this would be a one-person job, but he says he’s realizing it’s a lot more.

“Figuring out how to manage the personalities is a skill required in itself,” he says. “We spend a lot of time talking to people, ensuring that conversations are happening, monitoring closely how those relationships are going and resolving disagreements or misperceptions when they come up. We don’t want people to lose faith. They need to feel comfortable.”

Building the plane

Kansas City, Missouri, is eager to give the GovCombinator idea a try. Mayor Sly James and other top city officials not only like the concept, but say they want to see it housed right at City Hall. The main thing holding them back at the moment is office space. The plan is to convert an underutilized employee lounge into a co-working space, but they’re trying to find as much as $250,000 to do the renovation.

Kate Bender, with the city’s Office of Performance Management, says there are a number of public issues that would be well-suited to run through the Combinator. Many are related to transportation questions in the city center, such as ways to encourage biking and walking and how to handle the parking needs of a growing downtown population. Other candidates include a “smart city” plan to use sensors to collect data along a new streetcar line as well as some questions around implementing new development-permitting processes.

“These are ongoing projects that have a lot of stakeholders and cross-sector collaboration,” Bender says. “But there’s no physical space or methodology around convening them.”

Issue selection is an important part of the Combinator concept. A topic for a sprint would need to be high-profile enough to sustain the interest from a mayor and his or her deputies, as well as other stakeholders. But it also would have to be limited enough in scope to be solvable within the 100-day timeframe. “It’s not a lot of time,” Heller says. “It’s probably not the place to solve existential, generational problems like race relations. Although you could narrow that topic and maybe work on it in a particular neighborhood.”

Heller and Whittle imagine that a Combinator could have three sprints running simultaneously, each with a different cast of stakeholders. One topic would be chosen in a top-down fashion—whatever the mayor wants it to be. Another topic would be more bottom-up, coming from the public through a crowdsourcing mechanism. The third topic would come from the middle, via nominations put forward by city managers.

Still, there are lots of ways some of this could go wrong.

First, there’s scheduling. Would enough of the right people be able to vanish from their own offices, bosses, and colleagues to work out of the Combinator—at least part-time—for three months? Heller thinks getting buy-in from mayors’ offices upfront will be key to getting participants to show up.

Then there’s public perceptions. Would the public see the Combinator as just another back room where insiders strike deals? Heller says citizen voices would be ably represented through civil-society groups. But the question of who would watchdog work in the Combinator and how remains a bit fuzzy.

But the biggest question is whether the Combinator would really create better public policy. That would be almost impossible to measure. The real test is likely to be whether participants finishing a sprint feel it was a more efficient and effective way to get their work done.

Heller acknowledges that there are lots of unresolved questions. But his experience with the “unintentional coolness and magic that happens from co-working” suggests it’s worth finding out what the answers might be.

“We’re building the plane while we’re flying it,” Heller says. “None of us can guarantee it will work the way we want it to. I suspect two years from now it will look somewhat different, but it will be informed by experience and hopefully expertise. We’re lucky to have a shot at trying it.”

This story originally appeared on Citiscope, an Atlantic partner site