A crowdsourcing model for investigative projects

I left a great job with great friends in a city I love one month ago to work on an idea. I have this academic year as a Knight Fellow at Stanford to make it work. The clock is ticking.

So I decided to share my first, dirty prototype with the other fellows and readers of my own blog on Tumblr. As Stanford’s already drilling into my head, you need to prototype (and fail) early and often. I put it out early to get as much feedback and critiquing as possible from the jump. (Click here to read about some of the skepticism and enthusiasm that rolled in.) 

The goal

To create a model for an entirely community-driven, crowdsourced investigative project. Each project will be done along a strict two-year time frame and funded through nonprofit donations. It will also experiment with ways to compellingly present big-picture context. The users come along on a real-time quest narrative with an investigative reporter.

At the end of each two-year cycle, we wrap up our findings, create metrics for change and a guide for where to go if you want to continue following the story. Then we start over again with a totally new topic and raise the money for the next endeavor.

It’s kind of like ProPublica meets Ezra Klein with a touch of Planet Money. 

The first step

It all starts off with a simple call: I’m beginning a two-year exploration of (fill in the topic here — U.S. food policy, development in one San Diego neighborhood, U.S. foreign policy in Central America).

If you know something, help me out. If you want to know something, ask me. If you have just always wanted to follow this topic but couldn’t understand it in the blizzard of daily news, jump aboard. We start from scratch together. I’m not an expert.

Three quick steps after that:

  • We find and link to the best work that’s already been done.
  • List who you should follow and read on the topic.
  • Determine together the topic’s next frontier: What still needs to be explored. From there, we pick the focus for the first investigation.

Why these steps are vital 

They ensure that whatever work we undertake is completely unique and that excellent work is highlighted and not replicated. They make the process of story selection transparent and community-driven. It answers the question: What is the community need? Not: What does the reporter care to do next? And they give contributors three specific assignments from the start rather than just a nebulous call to action.

Important note on crowdsourcing 

This isn’t done exclusively through crowdsourcing. There is proactive traditional reporting being done at the same time. Interviews. Meetings. Reading documents. Developing sources. I’m a reporter, a community organizer and a community guide.

Once we pick a topic

The heavy lifting by the crowd will be the topic selection and the ideas that come with it. After that, it’s going to be vital for the reporter to continue to be a regular presence in people’s lives in order to keep them engaged and interested. I can’t just disappear for months. But I can keep them updated and also become a general guide for the issue. So throughout the reporting process, there will be some combination of the following:

  • Reporters Notebook: Interviews, photos and documents.
  • Explainers: Breakdown of interesting wrinkles, or issues from the broad topic that become big news.
  • Sending Out the Bat Signal: We’ll ask the crowd to help answer questions, look through documents and weigh in on decisions of direction.
  • Link of the Day: Best thing we’re reading, seeing, hearing or experiencing.
  • Weekly Aggregation: Top reads of the week on the topic.
  • Updates: Tell people where we are in the investigation and where we’re heading.

One question I’ll need to figure out: How often do I need to publish? Weekly? Daily? Daily seems like it could be too much of a burden. Is weekly enough? I don’t need rigid rules, but I will need some obligation to keep me focused. 

When the first investigative project Is done

The end project likely resembles the kind of collaborated, cross-platform investigative reporting project that ProPublica has pioneered. It’s a text story for a print and online partner. It’s a video story for Frontline. It’s a radio story for NPR. So, even if the crowd-sourcing kind of flopped, it would still be a high-impact traditional investigative project. 

At the end of two years, the whistle blows. We wrap up all of our findings. Everything we’ve learned. And we create a guide for users: here’s where to go and who to follow if you’re still interested. Here’s everything that’s changed. Here’s everything that’s wrong but hasn’t changed.

Potential for revenue

There is also a potential for revenue here, too. Maybe there are eBooks to be done on the topic we delved into, on our quest and even a journalism book on what we learned about harnessing the crowd or how to replicate. Maybe we do events throughout the project or at the end. Or even beginning.

Finally, we pick the next topic to tackle. We raise a new round of two-year funding commitments. And then we’re off.

Even if you’re not totally into the topic, maybe you’ve become connected to this community or to the reporter’s exploits — like a good detective show — and want to keep following along.

Two other important notes

1. Time: The entire structure will be built on two-year cycles. This provides the story arc with clear signposts — here’s where we start, here’s where we finish. Beats with expiration dates, as Josh Stearns called it. And it provides clear starts and stops for fundraising. More on that below.

2. Fresh Eyes: I’m coming to these topics as a non-expert. Eventually, I’ll get there. But I like the idea of (a) having a fresh set of eyes on a topic and (b) having a reporter know that the users likely know more.

Fundraising thoughts

This will be a non-profit endeavor. The two-year cycle, beyond its journalism purposes, also fits nicely with a specific funding problem: Often, funders want to build something new. Not give to ongoing commitments. In this way, every two years this organization or team completely changes. That opens the door specifically to raise money within the topic areas that we’re focusing on. So one cycle it’s organizations interested in food policy. In the next it’s urban planning. In the next its health care or the environment.

At least at the beginning, I’d be hoping for support from broad journalism foundations who’d want to invest in creating a template for a new, community-driven brand of investigative reporting.

Since we’re already harnessing the crowd, Kickstarter and other simple community-driven fundraising campaigns can be used. There could also be the potential for partner news organizations to share in some of the cost. Events and eBooks could provide revenue as well.


There’s another almost side project I want to tackle within this project: How to compellingly present context on the web.

How do we draw in readers of all levels of knowledge into a story together? How do we create context and explainers without them becoming (a) dated quickly and (b) sounding like term papers. Maybe we create different mini tutorials for people depending on their levels — novice, intermediate and expert. This idea needs to be developed more.

Final thought

My first instinct is to try to do this essentially as a one-person organization, albeit with guidance of colleagues and friends. However, the more I’ve thought and talked about it, the more I realize it is a project that can also fit well into an existing nonprofit organization.