If you hear “MVP” and the first things that come to mind are sports stars like LeBron James or Tom Brady — “most valuable players” — you’re not alone.
But in a growing number of city halls, another kind of MVP is making its way into the conversation. It’s the “minimal viable product,” and it’s a concept that comes to government from Silicon Valley rather than the sports world.
This kind of MVP is a critical piece of how entrepreneurs experiment, learn, and build products the market wants. One famous example comes from the online shoe store Zappos. In the beginning, the website offered only a few products and had no back-end behind it — when orders came in, the founder bought the shoes and shipped them out himself. It was a bare-bones operation that confirmed something nobody was sure of at the time: People were willing to buy shoes without trying them on first.
Now, city leaders are adapting this model to the public-sector context. MVPs are becoming a critical tool in the city innovator’s toolbox, one that helps them take a more experimental approach to building programs and addressing residents’ needs.
So, what’s a minimal viable product and how can city leaders use them? Bloomberg Cities breaks it all down here in our latest explainer.
What is a minimal viable product?
A minimal viable product is the simplest possible version of a program or service you’re developing. An MVP presents just enough of the underlying idea to begin testing out its main hypotheses with residents to generate feedback.
Harvard Business School Professor Mitch Weiss regularly leads mayors and senior city leaders in discussions about MVPs in a class he teaches through the Bloomberg Harvard City Leadership Initiative. He also writes about MVPs in his new book about public-sector entrepreneurship, “We The Possibility.” To understand MVPs, he said in a recent interview, it’s helpful to think about the top-down way that public-sector programs too often get built.
“Let’s say you want to start an afterschool tutoring program to teach kids math after they play basketball,” Weiss said. “You can imagine a mayor having the program built — they hire the tutors, build a curriculum, secure the space, create a schedule. And maybe kids show up, or maybe they don’t.”
Another way to go about it, Weiss continued, is to identify your biggest uncertainty — in this example, it might be whether kids will want to come for tutoring at all. “You could put up a sheet with the name of the program and an email address in the locker rooms and say, ‘Email here if you want to come for tutoring,’” Weiss said. “Now that’s not the tutoring program — it’s not even close to the tutoring program. But if you do it right, it will potentially answer the question — Are people interested in this? — before you build the whole thing. That little signup sheet is a very rudimentary minimal viable product.”
Why are MVPs useful in the public sector?
The whole point of using a minimal viable product is to learn, and to do it quickly without spending much money. It’s an idea that Sioux Falls, S.D., Mayor Paul TenHaken was familiar with in his previous career as the CEO of a digital marketing company, and now is working to deploy in City Hall. Using MVPs while developing new programs yields better services in the long run, he said, “because you’re not going all-in on an idea that you’re not sure if it’s going to work or not.”
TenHaken said Sioux Falls is now testing whether an on-demand transit service could fill gaps in the city’s patchy bus network. The experiment is limited to Sundays, when buses don’t normally run: Residents can request a ride, and a bus picks them up along with others nearby. Learning is the goal, TenHaken said — and one big lesson so far is that a lot of passengers aren’t comfortable using a smartphone app to request a ride.
“This is a more cost-effective way of doing government,” TenHaken said. “We could have said, let’s buy a fleet of 100 micro-buses because maybe that worked in another city so let’s do it here. And then when we start rolling it out if for whatever reason it just didn’t pick up here, we’d have this big investment that we’d have to eat the expense of or somehow figure out how to make it work.”
Is an MVP the same thing as a prototype? What about a pilot?
A minimal viable product is a close cousin to two other concepts that come up a lot in public-sector innovation: prototypes and pilots. All of them are tools for testing out ideas and learning in a low-stakes environment. But the concepts are similar enough that the terms often get used interchangeably. An MVP can look a lot like a pilot, albeit a super-stripped down version of one.
“It’s important not to get hung up on the lingo,” Weiss said. The point is to “maximize the amount of learning while spending the least amount of time and money.”
In what situations are MVPs most useful?
Sarita Nair, chief administrative officer in Albuquerque, N.M., is a big proponent of using experimental approaches in government and has trained hundreds of city staff in how to do it. She said MVPs are most useful when city leaders “are either trying to start something or change something.”
As an example, she cited a city initiative that aimed to get residents who were experiencing homelessness connected with the various city services. As part of that program, she thought it would be a good idea to put outreach coordinators onboard city buses, where some of the people they wanted to reach were known to spend time. The MVP version of that idea was to try it out one day on just one small stretch of a bus route.
“I mention that was my idea because we learned that it was a terrible idea,” Nair said. “People were less willing to talk to the outreach coordinators because there were people around who could hear — and for a few other reasons. So it was a really bad idea, but we were able to figure that out in one afternoon.” (The same experiment found that homeless residents welcomed these interactions once they were off the bus.)
What are challenges to using MVPs in government?
Minimal viable products often don’t work out as hoped — that’s not only expected but also how they save money in the long run. Still, that can make some people in the public sector uncomfortable. “There’s a little bit of risk in putting some resources toward something we know more often than not will fail,” TenHaken said. “We’ve got to get comfortable as mayors and as leaders knowing that the public actually does want to see us try and sometimes fail, versus not even try at all.”
The author Eric Ries, whose book “Lean Startup” popularized the idea of MVPs, acknowledges that putting an imperfect product out in the world is a difficult thing to do. “After all, the vision entrepreneurs keep in their heads is of a high-quality mainstream product that will change the world,” Ries wrote, “not one used by a small niche of people who are willing to give it a shot before it’s ready.”
Scaling an MVP — that is, figuring out how “minimum” the viable product can be — can be hard for local-government entrepreneurs, as well. Tim Fife, an innovation and strategy consultant and design educator, is coaching a city team in Bend, Ore., as part of a Bloomberg Philanthropies innovation training program. The team worked with residents to develop a simple idea for keeping sidewalks clear of snow: Anyone who needs help shoveling can put a flag out as a signal for neighbors willing to lend a hand.
For the MVP, the team had to resist adding additional layers to the idea — like using an app to call for help or turning shoveling into an opportunity to earn community-service hours. “We shaved all that off to get back down to the core concept of the flags,” Fife said. “You want to test the simplest possible version of the thing you’re trying to do. You don’t want to overbake it.”
How can city leaders overcome those challenges?
To gain comfort experimenting this way, Nair said it’s important to convey to city workers that they have “the organizational safety to try it.” TenHaken adds that clear communication with the public is critical. “What we’ve tried to do with our MVPs is just make sure the public knows we’re doing this, that we’re very transparent about it, and that there’s a good chance it won’t work — but that even if it doesn’t, we’re going to learn from it.”
Actively engaging residents is important as well. That’s as true for using an MVP as it is for the entire innovation process leading up to it. “If you try to do an MVP without having a co-design process sitting behind it, it’d be no wonder people would be worried about putting out a half-baked idea, because they’d lack the evidence needed to feel confident the idea is a good one,” Fife said. “The co-design process is a confidence-building exercise. You get the community to be part of understanding the issue, providing insight for the issue, inventing solutions to the issue, and then prototyping solutions to the issue. Residents need to be involved in all of that.”
Read more Bloomberg Cities explainers: