Managing an Open Data Program

Four principles for running a successful open data program are UI, UX, community, and challenges.

How to build a civic technology ecosystem: 4 Principles from Los Angeles

I’ve got four key principles, the first is focusing on the user.

1. Focus on the User

This is a picture of our first data portal from the city of Los Angeles.

This was launched in June 2014. I started a few months after the launch — in September — and I figured it was a good opportunity to get some feedback and do some user testing. I went around and started doing some basic user feedback. I just knocked on doors and asked people, “Does this site work for you? Are you able to get the data you need from this interface?”

It was interesting, very interesting, in fact, the feedback that I got the most kind of blew my mind. It was, “Abhi, yeah… kind of, but I don’t know what data is.” “What is data?” was the question that I got asked the most. Even someone in my office asked me that question.

“What is data?” was the question that I got asked the most.

What I decided then is we should think about interaction around open data that visualizes it compellingly, to let people really know what data is and how it affects them. Simply enough, we redesigned the portal a month ago, this is the new look of and just at the front it’s pretty basic.

We just put a map, we put points on a map that’s kind of beautiful, it’s Easter themed, because we’re coming up to Easter. It shows building permits that you can actually zoom in and find them near you. Scroll down, there’s search functionality, there’s discovery functionality like you see in most data portals, but the key feature was I wanted to say right there front and center, here’s the data that can matter to you.

We keep it relevant too; if there’s a particular event one week, like this last weekend in Los Angeles we had a big bike festival called CicLAvia, where they shut down parts of the street to let people just bike all along the street, so we put a map of the route front-and center. We made it relevant and visual, and here’s what happened when we did that.

The bounce rate, for anybody who knows technology, this is like if you go to a website and you don’t find what you need and you leave immediately, that’s called bouncing off the site. Our bounce rate before hand was 50%, half the people that came to went right away, not great for me. It dropped to 5%. Just by talking to people, figuring out what they wanted, what their problems were with the interface, had this kind of a dramatic shift in less than a month.

Just by talking to people, figuring out what they wanted, what their problems were with the interface, had this kind of a dramatic shift in less than a month.

I bring this up as we think about building open data experiences, if you think about building open data programs, keep in mind the user. Here’s just a preview by the way, of what the new data view site will be… right now it’s just spreadsheets on the internet, which it is. If you just go, it looks like a spreadsheet on the internet. This is what it’s going to look like soon, where every data set is going to be visualized and made interactive for people to explore and understand in a more meaningful way:

That’s the first principle, focus on the user. The next thing I think is, changing the citizen’s experience.

2. Change the Citizen Experience

We tend to think of open data as something that’s on a portal or maybe on an app that someone’s built, or maybe you download to your computer and do some research on. I think that’s just the start, because the open data that we’re providing can be really meaningful to private businesses, and the interactions you all are creating every day for citizens. There’s no reason you can’t put this data to work for your company. Here’s an example of that. The Atlantic wrote this article, How Yelp might help clean up the restaurant inspection industry. In the states, we have these things, these are restaurant inspection scores, do you have these in France? Yeah. Are they usually hidden in the back of the restaurant, where you can’t see them? Yeah.

This is a problem, right, because you want to get this information before you choose to go to the restaurant, let alone when you go into the restroom after you’ve ordered the main course. What we thought of, this is when I was back at Code for America was, can we take this data that typically is hidden, and put it into the palm of your hand?

Trulia is a very popular apartment search website, having recently moved to Los Angeles, I can tell you it’s very valuable, even though rents are really high. What they did is they took building inspection data; is your landlord a good landlord? Do they keep the building up to code? Are there termites, are there rats? The city is actually doing a lot of work to inspect that, they send inspectors to apartment buildings to see what’s happening, but again, that data, that information typically lives in some code enforcement, frankly piece of paper. If you’re lucky, excel page, if you’re really lucky, web page, but no one is lucky enough to actually have a citizen go to that web page and check it out before they pick an apartment.

I’m sure none of you have ever done that before picking a place to live. Now, any time you’re looking for a place in a place that has this open data, you get that right in front of you. For the private sector folks in the room, I would encourage you to think about this: Think about the work that you’re doing, and how you might be able to integrate open data into it, to improve your business operations, but also to improve the experience of the customers that you have. The broader point here is, instead of thinking about building more civic applications, which is what we tend to think about, how do we make all apps a bit more civic?

The third piece here is enabling community.

3. Enabling a Community

In my past role at Code for America, we spent a lot of time working with other cities to set up their open government, open data organizations. We worked with cities like Chicago, like New York, and like Boston, to actually develop their open data communities. I took some time, I sat down and said, “All right, what do we do, what worked, what didn’t work, and what might be a model for us to apply here in Los Angeles?”

The first thing you have to do is you have to create a place where people can work on good problems — where good people can work on good problems. This sounds simple, just like having a place where people can come together, but it’s very powerful, because what right now, what typically happens with open data is somebody will go into their basement with their computer, go to a web site and pull it down and work on it themselves. Maybe they’ll go to a coffee shop and meet with a couple of their friends and work on it together, but never do those communities get stitched together, unless you find a way for them to come together.

Second is building an identity. People like being part of a group, people like having their tribe. You need to create that sense of identity, and it’s code for Boston, there’s Beta NYC, it’s frankly as easy as creating a brand, but once you do that, you can start putting stickers in the back of laptops, as you might see for some people over here, you can start having common message boards and threads, you can just create a sense of all of these people are part of a common group, and that creates group cohesion. It’s not just random people meeting up randomly, you’re a part of this group.

4. Make a Challenge

Then, you’ve got a challenge of them to get something done, and here I’m talking to mostly the folks who work inside government. It’s us who see the really hard problems from the other side of the counter. It’s us who see the challenges that need fixing, that too often don’t get fixed, so we need to create mechanisms where we can give those ideas, give those opportunities to citizens who are interested in using our data, and say, “Love that you’re interested in our data, love that you’re interested in making our city better, here’s something you can do, here’s a hill that you can take.”

In Los Angeles last year, we did something like this, we had a #techLA competition. It was a 24 hour hackathon, and they built an app that made it easier to find homeless shelters, which I know this is something you’re passionate about, and the great thing is the winners of one more thing were all in high school.

It’s really people of all ages who are able to use this data to solve problems. This year, instead of just doing the one hackathon, as I was mentioning, because we want to make it a more of a community building effort, we’re doing a broader year long what we’re calling X Prize for LA, where we’re actually challenging people around key issue areas, for instance water conservation, which is a central challenge for us in Los Angeles, also immigration reform. Taking these big issues that we have and saying: “Hey citizens, you’re interested in working with us? Here are the things that we want you to do, and if you do it, we’re going to help you out. We’re going to give you resources, we might give you a city contract, we might get you a Hollywood star to do an ad for you, because we’re in Los Angeles and we can do that.”

And no matter your city, you can find important challenges residents can address with data and creative ways to reward them – that’s how you truly put data to work in your community. Through empowering the people in it.