What’s Next for Ballot.fyi

The simple site shouldn’t be on the role of citizens, but now it might have to be.

A few months before the election, I built a little site called ballot.fyi. Its goal was to break down local issues to get people interested in politics again. On the California ballot, the 17 state propositions included the death penalty, plastic bag ban, and weed legalization – just a few of the important issues that were getting little attention. To my surprise, the site reached a million people in its one month alive, gathered 300,000+ likes/shares/comments on Facebook, and had a ~37% return rate for visitors. (For those interested, I’ve included a lot more analytics info at the bottom of this article.) Overall, despite having been created and run by one person, people trusted and used the resource.

That was good, but oh yeah, also, the media snafu of the presidential election happened. I’m proposing that now, in this distrustful political era, citizen journalism might be crucial to educating the electorate.

But that’s a scary thought — citizen journalism? Isn’t that Twitter? Or Facebook rants that conflate facts with opinions with conspiracy theories? Isn’t that #fakenews? No, not quite. Journalism from citizens still has journalistic integrity, it still relies on facts and sources, and it makes it easier for the people to understand something complex. However, unlike traditional media, citizen journalism can take a different approach to rebuild trust.

Ballot.fyi took on citizen journalism in a new way. Here’s what I think worked with ballot.fyi, and why this kind of media must continue into the next elections.

Being nonpartisan empowered the reader

Ballot.fyi’s headline was “The quickest, nonpartisan guide you’ll find.” It’s the “nonpartisan” part that lowered the barrier to sharing it socially. People didn’t have to stick their neck out by sharing it. They could instead simply tell people to vote in an informed way, and remind people of their civic duty (while still complaining about the number of propositions).

People generally repeated the same message the site had, and their trust that it was nonpartisan lead to more sharing. (I know none of these folks)

More importantly, by being nonpartisan, people reached their own conclusions, which (I would guess) resulted in a much more confident and robust opinion.

I achieved (more appropriately: I approached) nonpartisanship by thoroughly researching both sides’ claims, presenting at least one counterargument to every argument, citing every source, and then having both liberal and conservative friends edit the text. Was it completely unbiased? It turns out some blogs thought I skewed conservative, while some Reddit users couldn’t bear my liberal slant. People from both sides of a proposition would email me, explaining how my summaries are biased one way or the other. If the timing was right, I’d sometimes copy/paste the main points of someone’s email and then send it to the opposition to see how they’d respond. I worked hard to make the summaries as nonpartisan and true to fact as possible. The site loudly solicited fact-checking and corrections — I truly think that having the door open for feedback helped build trust with the reader.

Let’s be real. It’s not vogue to admit it, but nonpartisan media exists. Unfortunately, even the most innocuous of them are still seen as Media, or worse, Mainstream Media. And sadly, to many Americans, that now automatically triggers a heavy skepticism. At best, to liberals, it’s a light skepticism. When the media is not trusted, we have no recourse but to spread real information through the channels where it is trusted.

Rather than make a boring list of ‘for’ and ‘against’ arguments, arguments are structured in a conversation to allow for counter-counter-counterpoints. This format is an example of something ballot.fyi could do that traditional publications would not be “allowed” to do.

Maximizing quality over scale

The goal of ballot.fyi was to make voters feel smart, establishing trust, while feeling home-grown. We maximized quality while others went for scale. This home-grown feel showed that we had no ulterior motives (we didn’t).

This was not a platform, nor even a business. It didn’t have any ads, user login, crazy swipe interactions, or a dizzying site structure. In fact, except for the landing page, ballot.fyi was just words. That’s it. Good, high-quality words. People went into it expecting to learn more about each proposition, nothing more.

Ballot.fyi wasn’t the only voter education tool. Initiatives with more money and manpower — Facebook, Voteplz, and BallotReady (to name a few)––pulled from Google Civic API or Ballotpedia to educate voters. This was reasonable. Their audience and strategy was national, but it resulted in the same, dry information that you’d get from the government (or an API), and so the tone was very impersonal. The government has to be impersonal; these sites do not. To appeal to a younger section of the electorate, ballot.fyi swooped in with a different tone.

Ballot.fyi was California-only, but that made it easier to create a tone and brand that made sense for California. We were thorough yet approachable, cheeky but objective. It hit a tone that neither journalism nor government are allowed to take, but it built trust and further distanced it from the now demonized mainstream media.

A nonpartisan + personal brand is difficult to hit, but it might be a way forward for citizen journalism to get good information to people.

Up next — expanding ballot.fyi and more citizen journalism

The mission for ballot.fyi will remain narrow — get people excited about local issues and local politics through the ballot. With some funding, we can help activate and form local chapters of content-writers for a florida.ballot.fyi, for example. Ideally, this would look and feel different from an ohio.ballot.fyi. By the 2018 elections, we hope to have all the swing states, their propositions and elected officials. By 2020, every state and most large cities. It will be slow, but a bottom-up approach, through citizen journalism, seems necessary.

Perhaps the larger mission is to create a movement of citizen journalism to reach a broader audience. People want information. It may not be called news anymore, but people will always want to feel smart, and it’s up to almost everybody now to start to create and vet citizen journalism with integrity.

Stay tuned.

Appendix

The performance of ballot.fyi

All stats are from Oct 3, 2016 to Nov 12, 2016; tracked and generated by Google Analytics.

It went viral within a day of posting about it, stayed steady at 20–30k sessions/day until Election Day, where it peaked.

All:
Unique visitors: 994,314
Pageviews: 7,508,350
Average time spent: 7 min, 38 sec
36.2% of all 1,558,685 sessions were returning visitors

Subset B: visitors from a California location:
Unique visitors: 853,981
Pageviews: 6,802,817
Average time spent: 8 min, 01 sec
37.2% of all 1,369,489 sessions were returning visitors

These are probably the folks who used this to vote:
Subset C: visitors from a CA location who spent at least 10 seconds
Unique visitors: 540,568
Pageviews: 6,423,647
Average time spent: 10 min, 58 sec
45.4% of all 1,000,393 sessions were returning visitors

There were 309,861 sessions that were greater than 10 minutes.

In Subset C, this is what the age breakdown looked like:

Pretty much on point.
Also, in Subset C, most people got to the site through social links (Facebook more than Twitter)
The site reached towns large and small, but in the end, LA beat the Bay Area

Friends of ballot.fyi

This article is, in part, a post-mortem of ballot.fyi. As such, I want to thank the many who helped make it what it was: Rocio Reyes-Morales, Bailey Cypheridge, David Murdter, Mirage Marrou, Christine Takaichi, Matt Crowley, Ehrik Aldana, Alastair Warren, Elliot Lynde, Maddie Boyd, Nick Inzucchi, Michelle Fang, Benji Renzo Kuroda, Christine Hendrickson, Nivi Ramesh, Andrew Noyes, Aviva Rosman, Fouad Matin, Hunter Scarborough, and everyone who shared the site in any way.


Jimmy Chion, an artist and engineer, has spent a lot of time thinking about how to show people complex things in simplified ways. His most recent project, ballot.fyi, helped educate people on the local election. He’s previously been an Artist-in-Residence at Autodesk, Pier 9, and a designer at IDEO.