Paul Bradshaw is an investigative journalist and author based in Birmingham, U.K., who teaches online journalism at Birmingham City University and is also a visiting professor at City University in London. He publishes the Online Journalism Blog and recently wrote about data visualization for The Guardian’s Data Blog. CJR assistant editor Lauren Kirchner spoke with Bradshaw about Help Me Investigate, a project he launched last summer that invites readers to suggest local governance issues for each other to investigate—issues that the mainstream media might miss. This is an edited transcript of that conversation.
What was your motivation in starting Help Me Investigate?
I was researching crowdsourced journalism for a book chapter, and a couple of examples I found made me quite excited about the potential for engagement with local issues. For instance, The Florida News Press really brought it home [with their crowdsourced investigation into price-gouging by public utility companies] with how popular that story was on the website; that would have been quite a dry story in print, and probably not that widely read. But more broadly, I could see where things were going. I’ve been writing about online journalism and technology for almost ten years, and I’d gotten to the point where I could see that…if you’re going to choose where to invest your efforts, investigative journalism seemed to be area that most needed some support, most needed a way to make it effective.
With a crowdsourcing approach, you’re not just telling a story; you’re actually passing along information and expertise to people, so that those people can then help you in the future. So the fact that I’ve been involved in investigations on the site, means that I now know about parking laws, or the deregulation of transport routes, and so I can help with future investigations [on those issues]. I think that’s a key part of crowdsourcing. It’s about diffusing expertise as well as just finding and telling stories. You’re empowering people to become part of your team.
What types of stories did you think were the most important to investigate?
There’s a lot of academic literature out there about news values and gatekeeping, and what is seen as ‘newsworthy,’ and so on. One of the things I wanted to see was, if you opened up the editorial process and handed off the agenda to users, would you get a different mix of stories? In the 1980s there were so many ‘miscarriage of justice’ stories, but for some reason you hardly ever see those these days; they’ve fallen out of fashion. Whether something gets investigated or not in traditional media depends on all kinds of things, like whether it’s a fashionable subject, who’s being affected by it, whether that audience is going to be attractive to advertisers, whether the legal risks are worth it commercially, all those kinds of things. Well, you don’t necessarily have those same considerations in a crowdsourcing exercises.
All of the investigations we do are chosen by the users of the site, it’s not me choosing what to investigate. And some of my favorite investigations have been the very small, personal investigations that you would never get in a local newspaper. One person asked why there were two bus companies running buses on the same route. Someone else asked how they could find out who owned a particular piece of land in Birmingham. These are really specific investigations, which you might see as too trivial, but they are actually quite important to those individuals, and can lead to other kinds of civic engagement, and, more broadly, people holding power to account.
What is the typical process of an investigation on Help Me Investigate?
Anyone who is a member of the site can start an investigation; an investigation is basically just a question. For instance, ‘What is happening with the Birmingham Council website?’ or ‘What is the worst street in Birmingham for parking fines?’ Then we give them advice, on two fronts: first of all, how they might begin to investigate it or find more information, and then how to build a community around that question. So there’s a social task and there’s an editorial task there, effectively. Then we’ll break down the main question into smaller tasks: they’re called ‘challenges.’ A challenge in investigating a business might be to find a particular year’s accounts of that business, or speak to a particular person in the company, or look at [public] reports. So the question is broken down that way. Anyone can join on to an investigation, and we can invite people with specific skills who we think might be able to help. And as the site builds up, more and more resources start to build up, and we build a bigger network of support.
Do you ever hand off a story that’s gotten really big to the mainstream news?
Yes, quite often if a story gets big, we might assign the challenge of writing up what we’ve found as a story, so that other people can get involved, and at that time we’d probably invite a journalist from a local paper, or anyone else who was interested. Our story on parking fines, for example, was reported on by the local newspaper in Birmingham. Or sometimes if there’s a national investigation [reported on in the mainstream media], we might take the story and do something local with it.
Do you see the Help Me Investigate model as replacing a traditional newsroom, or supplementing it? If you can imagine this model spreading to other towns and cities, what do you think that would look like on a larger scale?
I certainly see it as complementing mainstream journalism, and journalists who have the kind of community management skills to get involved in it could really benefit. I think the main thing that’s stopping us at this point is the cultural barriers, and the time available. I’ve had quite a few conversations with individual journalists, and there’s a will there to get involved with this sort of tool, but quite often there just isn’t the investment of time or the cultural understanding to be able to pull this off, I think. They’re quite happy to take the stories at the end, [laughs] but I think community management is a rare skill in print newsrooms, certainly.
But further down the line, I think this is going to become more and more common as the culture becomes more widespread, and I also think that we’ll learn better what works in time. It’s only a year into this project and I’m still learning what makes for a successful crowdsourcing project, how these communities work, what the dynamics are. That’s been my intention since the beginning: even if this project doesn’t work, we’ll learn a lot about how to do it better in the future.
What are the limitations of this kind of model? Are there certain stories that you just can’t do that might be more appropriate for a larger newsroom with a full-time staff?
I think that’s actually the smallest problem. The major problem is legal. There are certain investigations that are just too legally risky to do in public. One of the things I’d like to do is to make it open source, to allow people to do it in a distributive fashion and distribute the risk that way, so maybe people would be braver about taking on those risks. Having all the information there on one site, it’s easy for us to be silenced—just from the financial risk, rather than the actual truth of the investigation or anything like that. So the legal thing is probably the main thing. And then there are many types of investigations that are difficult to do in public: anything to do with the Secret Service, anything where the only way to investigate the issue is to have personal contacts inside an organization, and very close relationships of trust. That’s unlikely to take place on a semi-public website. But what does suit it is, if you’ve got investigations that need a diverse range of expertise, and perhaps have a lot of small tasks that can be completed by different people, those sorts of things suit this model very well.
Update: Paul Bradshaw’s bio above did not include his position at London’s City University in London when this post was first published; the omission has been corrected.