Walk into an annual strategy meeting of any think tank anywhere in the world and you’ll observe a tension at the core of almost every discussion. It can be summed up in a single question: What is our audience?
Some think tanks have clearly defined ideologies. Others are issue-driven. Still others aim to be explicitly non-partisan and opportunistic in how they choose the issues they engage in. But what all think tanks have in common is the mission to bring about social change through 1) the analysis of the effects of public policies and 2) the advocacy of new policies and practices that better meet the needs of a group of users.
Which brings us back to the question of audience. Are think tanks most effective when they focus their efforts on influencing elite decision makers, or when they aim to shift public opinion? There is a temptation to say, “of course it’s both and not either/or,” but each strategy has distinct implications in terms of staffing, communication, and branding.
They Don’t Care as Much as You Think
Let me start with a few untested assumptions about society at large:
- First, everyone values information that is directly relevant to their well-being. For example: How much will it cost to purchase a house in my city? What is the safety record of my car? What is the best hospital in my neighborhood? What movie should I see at the nearest movie theater? How much will I need to save for retirement?
- Second, everyone is interested in a few select political issues that are directly relevant to their lives. Farmers are interested in farm subsidies, but most people aren’t. Environmentalists are passionate about industrial contamination disclosure. Factory workers are vested in policy discussions related to outsourcing. Immigrants are passionate about immigration reform.
- Finally, there is a far smaller bucket of the population (less than 5% I’d venture) that is interested in just about every social issue, but also those root, underlying issues, such as campaign financing, lobbying regulations, asset declarations and public spending disclosure. That 5% of society tends to make up nearly 100% of those who work in think tanks, and they tend to vastly over-estimate the larger public’s interest in the issues that are of interest to them.
Sunlight Foundation, Aaron Swartz and Public Opinion
Sunlight Foundation Co-Founder Ellen Miller
When Ellen Miller co-founded Sunlight Foundation in 2006 she wanted to break free from the closed circle of DC insiders and get the rest of America involved reforming Congress, and she wanted to do so by taking advantage of new technologies. This made all the sense in the world; it was a sea change in terms of how people were interacting with information in the United States. Ever since 2006, Americans have been spending less time with newspapers, television, and radio — and far more time with their computers, smart phones and tablets. Sunlight Foundation has always been clear about its audience. It doesn’t aim to influence the decision makers; rather, it aims to shift public opinion so that the decision makers feel pressured to change, or else be removed by voters.
Over the past seven years the Sunlight Foundation has developed truly innovative and sophisticated technological platforms. Try out Inbox Influence, which will automatically show you the political contributions of the people and organizations that are mentioned in emails you receive. Or, along similar lines, Checking Influence will examine your latest bank statement and show you the political contributions of the companies you support as a customer. Sunlight Foundation has developed beautiful tools that help you track the influence of foreign lobbyists, view deleted tweets by U.S. politicians, learn more about political fundraising, and subscribe to customized alerts as legislation is proposed.
Over the past two weeks my indirect contributions (via my purchases) to Republicans and Democrats have been split even. eBay (which also happens to be the source of my salary) gives slightly more to Republicans while Apple and Netflix give more to Democrats.
Longtime employees of the Sunlight Foundation will be among the first to tell you that they have been disappointed by the overall number of users that their sophisticated tools have attracted. While so much nonprofit technology is hardly usable, Sunlight Foundation’s tools are beautifully designed, fully functional, and interesting. Yet they have only managed to attract a niche community of users.
This is something that Aaron Swartz (RIP) foresaw back in 2006 when he was invited to a Sunlight Foundation gathering in San Francisco:
As a Web 2.0 developer, it’s hard for me to see how even the best Web 2.0 site can have much of a positive impact on government. Genuinely promoting transparency requires the hard work of doing investigative research, publishing reports, and promoting them to the media. Bubble 2.0 hype aside, the fanciest pop-up windows and and Google Maps mashups won’t change that.
The attendees seemed to begin to recognize this … We decided our saviors would be the “Paul Reveres” — the people who care enough about politics to slog through the data and then mass email their friends when they find something good (we concluded that going after newspaper reporters was too Web 1.0). They would save us from having to write reports or take positions; all we had to do was make the data available and let them do the rest.
I’m sure there are a handful of people who actually do this, but it seems like we’re spending an awful lot to build a site just for them. And even then, what impact will they have? Even if our Paul Revere finds the smokingest of smoking guns and posts it on their extremely popular blog, without a larger political platform it will only fuel the cynicism that [Sunlight Foundation co-founder] Klein claims he’s trying to combat. (“There they go again,” the reader thinks, and hits the back button.)
Three years later, in a controversial essay titled “Against Transparency” for the New Republic, Lawrence Lessig built on Swartz’s thesis, arguing that many Sunlight Foundation projects do little more than generate public outrage over relatively small acts of malfeasance at the expense of working with decision makers to bring about major structural reforms, such as publicly financed elections.
There’s that same tension again. Should advocates that want a more functional, accountable US Congress work closely with the powerful decision makers, or should they aim to sway public opinion? Is it possible to do both at the same time?
Sitegeist and Human-Centered Design
In 2010 Sunlight Foundation received a grant from the Knight Foundation to develop a series of “National Data Apps” to help US residents “easily use federal data to better understand everything from local pollution and medical care to personal financial services.” The first app, Sunlight Health, was released in 2011 to relatively little fanfare. So far it hasn’t received any ratings in the iPhone app store, only one comment, and hasn’t been updated since 2011. The next app in the series, Upwardly Mobile, was released in early 2012 to allow users to compare indicators based salary, living and employment data in order to choose where they want to live. I’m not sure how many monthly visitors the site has (compete.com says it’s too low to track), but a search on backtweets.com shows that there have been no recent Twitter mentions of the site.
Then, for the third app in the series, Knight Foundation partnered Sunlight Foundation with the design firm IDEO to create an application that truly meets the information needs of its users. The result is Sitegeist, a beautiful, intuitive application for iOS and Android that automatically detects your location and then provides you with a wealth of relevant information including age distribution, weather history, average rent, average commute times, and housing statistics. Sitegeist is a departure from other Sunlight Foundation applications. Rather than seeking accountability in Congress, it aims to provide users with information that is relevant to their lives. The results have been impressive. Within the first week the app was downloaded 20,000 times and brought about 300,000 page views. It has been featured in the Washington Post, Boing Boing, Gizmodo, Fast Company, Good, and GigaOm.
It is paradoxical that Sitegeist’s success has depended in large part on Sunlight Foundation putting its own objectives to the side in order to focus on the needs and interests of its users. There is a term for this, human-centered design. (Or user-centered design, or “design thinking,” depending on who you’re speaking with.) Rather than starting with an objective, one begins by observing how target users interact with the world around them and how design can alter those interactions to lead toward a desired social change.
The human-centered design approach began with a focus on product and interface design, though a few innovative individuals are bringing the principles of human-centered design to the development of public services and public policies. At the risk of what surely comes across as mutually self-assured flattery, I highly recommend the work of Reboot, a design agency focused on international development. Later this year Stanford’s Design School will launch the Governance Collaboratory, a class that will explore human-centered design as it relates to governance.
Sitegeist as a Civic Entry Drug
The great untested hypothesis is whether an application like Sitegeist that provides its users with information that is of direct relevance to their lives can also lead them down a path toward demanding greater accountability from their government. It is a crucial question for my own work. Can, for example, IMCO develop a Yelp-like website that provides parents with the relevant information they are seeking about their children’s schools while also seeking greater accountability in the public education sector? Is it possible for an application to both inform users about their public transportation options while also nudging them to become involved in the advocacy for more sustainable, inclusive transportation policies?
My hunch is that, without focusing on the needs of users of these applications, no civic application will reach scale. On the other hand, I’m not yet convinced that those applications that have managed to attract significant attention are able to nudge their users to demand more of their governments.
Thanks for this thoughtful post, David. I need to quibble with a few things, though. I’ve been with Sunlight for for four and a half years, which I hope qualifies me as a longtime employee. And though I’m always eager to get more people using our stuff, I worry about the impression you’ve given about the reach of our work. Sitegeist is off to a great start, and we’re really proud of it, but it’s certainly not our biggest success. Our Congress for Android mobile app (iOS counterpart coming soon!) has been installed hundreds of thousands of times. We served over 300 million API requests in 2012, powering users ranging from Obama for America to the Tea Party Patriots (and the Wikipedia SOPA blackout!). Longtime Sunlight grantee OpenCongress serves millions of users every year. And media sources like NPR, Bill Moyers and the New York Times have used our data to power their reporting, reaching huge numbers of people.
Now, it’s certainly true that sometimes things we do fall flat. And sometimes projects are more about using technical novelty to attract attention than about making a sustainable tool. I mean: no one here really thought that the average Gmail user would want campaign finance data included in their everyday correspondence over the long term. That’d be nuts. But we were able to deliver a technical experience that was cool enough to approximate magic, and in the process get a bunch of reporters and citizens to think and talk about money’s effect on our politics. That was the game plan from the start.
Besides, not every problem calls for huge popular action (though some certainly do!). We’re tremendously proud of Scout, but we know its impact is likely to take shape through advocacy professionals, in stories like this one. That’s OK! If we can get the right information to the right people, that’s good enough. Democracy works, in part, through division of labor. We want to make it easier for people to engage with the issues they care about, but our goal is better outcomes and governance, not to turn every citizen into a congressional watchdog just for the sake of doing so.
Let me also add that, although I have tremendous respect for both Aaron Swartz and Prof. Lessig — it’s safe to call them both heroes of mine — I have found the depth of their thinking on transparency issues to be disappointing. Defining counterfactuals is always difficult, but there are good reasons for thinking that transparency’s value is less about catching bad action after the fact (a mental model of action that your quote from Aaron exemplifies well) than it is about forestalling bad action at the margin. I’ve written about this from a personal perspective here, though I hasten to add that I am not speaking for my colleagues in that brief essay.
With all of this said: you’re absolutely right that Sitegeist has proven to be a big hit for us, and I agree with the reasons you point to as the basis for its success. We’re anxious to learn from and build on that success! But while “reaching scale” for an app can open exciting possibilities, we don’t think it’s an end in itself.
Thanks for the thoughtful response, Tom. I absolutely didn’t intend to imply that “reaching scale” should be the objective of any of Sunlight Foundation’s interventions. In fact, an unstated hypothesis was that in some cases it might be most effective for an organization to focus its efforts on changing the opinions of specific decision makers rather than attracting mass public participation. One example is the State Integrity Investigation by CPI, Global Integrity and PRI. It’s about as low tech as you can get and hasn’t “reached the masses” despite some public radio coverage. But it was enough to put sufficient pressure on state governments to bring about some significant reforms.
Another hypothesis I have, which I think is supported by your comment, is that information related to government accountability depends on intermediaries (media, academia, political parties, issue-based advocacy groups) while information of direct relevance to the user’s life can reach them directly from the database to the interface.
Another way to put it: Despite our greatest hopes, it seems that just about no one pulls out their smartphone and ponders, “How can I hold my government to account today?”
Excellent discussion–
‘Despite our greatest hopes, it seems that just about no one pulls out their smartphone and ponders, “How can I hold my government to account today?”’
True, but I wonder if–and I say this at the risk of sounding techno-utopian–there is a middle ground. Perhaps the interactions prescribed by mobile devices are just flawed. Or at least flawed for this purpose (smartphones seem to be great for making phone calls.)
What if my phone rang when I reached a public park? On the screen would be recent statistics on city funding for that park. If I was not happy with what I was seeing in terms of the playground, benches, etc., I am now in the proper “headspace” to want to act on this *now*–could my phone permit me to engage in holding government to account for that particular issue at that particular time, given my physical/mental context?
I worry that the topics you’re imaging people wanting to engage with over a mobile device are too broad/abstract. There’s a scale mismatch and a channel mismatch. Maybe even a temporal one. Mobile devices are locative and for small bits of information–not the proper channel for a “congressional budget debate”.
People in the marketing world are, of course, wrestling with some of the same issues–how can I determine intent? How can I find the proper time to message somebody about my product when they’re in the buying mood? Google promised (and arguably delivered) this with AdSense because they could integrate multiple, (then) disperate channels of information to determine that hidden behavior, and deliver a message in a proper form (not intrusive, subtle text put next to everyday content like E-mail).
We need to do the same thing with civic engagement–reach people with appropriate “messages” on appropriate “channels” at the proper moment, when they’re in the “buying mood”. Getting people to act is tough. Again, marketers know this. But the apparent emotional payoff for buying a new pair of Nikes is a lot simpler to communicate than fixing the federal budget.