Where is the "Killer App" of politics?

The close of 2018 marks 18 months since I started working in the “political tech” space: via a series of small volunteer projects, with Amplify (an app that Indivisible groups use to coordinate action) and currently as a designer at Hustle. Coming off my time at Quora, I was excited to take what I knew about consumer product development and apply it to the political space ahead of “the most important election of our lifetimes”.

In that time, I’ve learned that the political space has distinctive properties that make it hard to run a traditional “consumer tech playbook.” In this post I’ll explore some of those properties, and the products I’ve seen doing the most interesting work in this space.

But first, what do I mean by the “Consumer Tech Playbook”? In short, I think the best consumer-facing products are built by:

• Focusing on a core activity that people love to do repeatedly and frequently (e.g. posting photos on Instagram)
• Tracking key engagement metrics
• Developing hypotheses for what will get people to do more of that activity
• Running experiments to test a hypothesis and learn what moves key metrics
• Launching enough successful experiments to achieve exponential growth in user engagement

At a high level of abstraction, this is how every successful consumer internet product started and grew.

Yet despite significant investment in the political space in recent years (and especially since November 2016), no consumer-facing product focused on political or issue activism has achieved breakout success.

To be clear, there are numerous ambitious and exciting pieces of political technology being built, with enormous amounts of impact at many parts of the political process. The Higher Ground Labs portfolio is full of these companies, and they just published an amazing overview showing just how much energy and innovation is happening in this space:

Without taking anything away from all the incredible work of these companies, I’m interested in exploring why so few look like the breakout consumer successes of the past decade — Facebook, Twitter, Instagram, Pinterest, etc.

Here are my theories for why:

Politics is cyclical.

Most people’s attention and interest in politics, especially since 2016, has focused on the Federal level. (Side note: This is a shame, as there’s an immense amount of impact to be had at the state and local levels.) Federal elections only happen every 2 years, and within that 2 year cycle, much of the energy happens in the final weeks before the election:

‍

So that leaves ~20–22 months between elections when most regular people aren’t paying too much attention to politics. This fallow period makes it challenging to build and grow a product loop, for a few reasons:

The biggest problem is there’s just not enough for people to do. Even the most politically engaged people don’t have many outlets for their passions during off-cycle years. The two best options, and those that have spawned the most successful political technology, are contacting legislators (5calls, Indivisible, Amplify) and donating money (ActBlue, Swing Left, Flippable). While these are highly influential and very worth doing, they don’t provide the addictive feedback loop that taking action on Instagram / Twitter / etc provide. An ideal feedback loop has a clear connection between your action and the feedback, and feedback which comes quickly. Neither calling legislators nor donating provides clear feedback (“Did my call convince Diane Feinstein to be tougher on this issue?”) and even when there is, it’s often delayed by weeks or months.

One of the best approaches to this problem I’ve seen is MobilizeAmerica, who aggregate opportunities across many issues and organizations, and provide many ways to have impact: Phone banking, text banking, writing postcards, event attendance, voter registrations, community meetings, and more. I think any successful product will need to have this diversity of opportunity for action.

‍

Almost as bad as the lack of engaging loops for users is the lack of good learning feedback for the product team. If a company is looking to take a data- and hypothesis-driven approach to product development, they need a data set of users and actions within the system to start running experiments. The paucity in off-cycle political actions results in precious few shots to test theories for increasing engagement.

At Amplify we had countless ideas for how to increase the number of people taking political actions: social feedback from likeminded activists, group challenges and achievements, identifying politically inactive friends to engage, etc. Like any brainstormed list, most of these ideas wouldn’t have worked, and hopefully some would have. But with the long feedback cycle of a small user base and months until key elections, we didn’t have nearly the opportunity to test our hypotheses. This was especially confounding as we had decent usage for a small political startup: 3000 MAU, using the app to call their legislators, plan meetings, and send postcards to voters in special elections. Yet even this wasn’t enough to give us the statistical power to do true data-driven product development. (It did make us jealous of AirBnb, who were running 500 concurrent experiments in 2014).

‍

Political talk is awkward.

Most of the successful consumer-facing social products of the past decade have had very simple mechanics (Like, Comment, Retweet) and content which was public by default (Twitter, Instagram, Reddit, Quora, Pinterest). These decisions have many benefits, both in simplicity of designing the system’s dynamics, and in the growth potential of each additional piece of content.

Unfortunately, politics is a notoriously touchy subject. Especially in our post-2016, highly tribal, everything-is-political world, it’s difficult to have constructive and emotionally safe conversations about politics. (As evidence, witness the cottage industry of articles about how to talk with your family about politics.) What does this mean for political technology? Designing a consumer-facing product centered around political speech and action requires an immense amount of subtlety in system design.

Here’s an example from one of the most interesting political apps I’ve seen, Vote With Me: You upload your phone book, and the app scans the voter files to identify contacts who haven’t voted recently, then has you encourage them to vote:

If a product can help me get Friend #3 to vote in the midterms, it has had incredible impact. I’ve tried to build a product that does exactly this, and I can say confidently that Vote With Me has excellent execution on this idea.

And yet… there’s still a long journey to make and grow the impact of this interaction: In my example above, the data isn’t great. I know that Friend #1 hasn’t lived in Illinois for years, so now I’m wary of whether Friend #3 really is registered in Colorado. Even if they are, I haven’t talked to Friend #3 for years, so texting them out of the blue and asking them to vote is a very strange thing to do. The suggested text is pleasantly non-accusatory, but I’m not sure I’d be brave enough to do it. And even if I did text them, and they apologized for their lack of civic participation and promised to vote in 2018, how can I build off that to generate a viral growth loop in my product? It’s unlikely that I could make a public feed saying “Joel convinced Friend #3 to vote even though they hadn’t since 2012” (public shaming isn’t an ideal foundation for a product’s primary content) which means no positive social reinforcement for all the work I just went through. I love what Vote With Me is doing, but don’t envy them the challenge of scaling this product experience.

Despite these social hurdles to climb, Vote With Me has achieved early success — boosting trips to the polls in the Conor Lamb special election by 2.4%. Experiences like this give me optimism that excellent execution and focus can forge a path through the social challenges of politics — but it requires a deft and empathetic touch.

‍

Grassroots organizing is hard work.

Experienced campaigners (of whom I am eternally in awe) talk frequently of a Pyramid of Engagement, the process by which people begin to participate in activism:

One vision for a killer political app would be “Make it easier for anyone to climb the pyramid of engagement on their own.” This could look something like: Expose people to issues they care about, develop associations with organizations they believe in, give an easy way to take lightweight action, start to take a more active leadership role, evangelize the cause to their communities, and beyond.

The problem: This is godamn hard to do! Community organizers have been developing techniques and playbooks for decades on how to do this effectively, and it requires a huge amount of knowledge and intimate human conversation. When I’ve modeled out how a product architecture might begin to capture this journey, it’s daunting: Each step of the pyramid is named and described; Users achieve a certain status by completing key actions and publicly move up the pyramid; Upon promotion, they are granted permissions to engage those lower on the pyramid; etc. This begins to feel like a didactic and top-down imposition of what is normally a fluid and human-driven process.

I’ve come to believe that a much more effective technique is to empower the organizers in executing their existing playbooks. This is one of the reasons I’m such a believer in the potential of Hustle to affect true change in the world. (Sorry for the self-promotion, I promise it will be over soon.) Our texting communication tool helps our clients (like Planned Parenthood, Sierra Club, Susan G. Komen) do the work they were already doing, by helping build the relationships that power movements. We fundamentally believe that organizations best know their own issues, communities, and plan for impact. The product we build helps them reach their goals by empowering organizers and volunteers in their conversations with people — rather than imposing our own view of the right way to do this work.

I’m hopeful that a version of“self-serve community organizing” can one day inspire new activists with no intervention needed — and I’ll be very impressed when a team figures out how to unlock this potential!

In 18 months of working in this space, I have been inspired by the work and innovation done by countless people in the “civic tech” world, which has been thriving for many years before I got here. The list above is by no means complete, and nothing articulated is an impossible challenge. I hope this post is read with the intention it was written: to start a conversation with other technologists from the consumer world about how to best use our skills to make change in the world.

‍