The future of online investigation?

O Conner
11 June 2010

Investigative journalism is no longer just the domain of the traditional media.

Newspaper sales are declining drastically and so (inevitably) are the staff. The Mirror group have recently announced one of the largest redundancy programmes of any news group, axing 200 jobs. As the number of reporters and journalists decrease there are greater time pressures on those that remain to fill up a paper with content. The result is usually an abundance of PR stories and a increasing dependence on news-wire stories that are often unchecked and sent straight to press.

The blame for these job cuts is normally focused on a loss of newspaper revenue caused by the free content available on the internet. But if the internet is the culprit for a decline in traditional investigative journalism, it is also breeding a new form of investigation.

For those that don’t know, crowdsourcing is the process of harnessing the power of the masses to accomplish tasks. By calling out to an interested crowd you can get tasks accomplished, ideas generated and develop the kind of insights never before possible.

When this technique is applied to investigative journalism you end up with a massive team of investigators. The most popular example is the use of crowdsourcing to investigate MP’s expenses – the Guardian has so far recruited 26,763 people to review MP’s expense documents.

Developments in collaborative online investigations are being pioneered by Paul Bradshaw on his website Help Me Investigate. Designed to connect, mobilise and uncover – it allows users to investigate “things”. These things can be anything from how much donation websites take from charities to issues surrounding the digital economy bill.

Anyone can suggest an investigation – an activist, journalist or a member of the public. The investigation then becomes a series of tasks. For example, one investigation asks “How orchestrated or organised was the #janmoir campaign?” and the tasks are:

  • To provide background information
  • Analyse tweets
  • Suggest ways to test “organisation” and “orchestration”
  • Compare it to other “outrages”
  • Follow the source that led to the outrage
  • Invite experts to take part

Other investigations can involve writing freedom of information requests, contacting local councils and identifying possible contributors.

Although this project has proved its worth on a local level – it is clear the vision of this site is for national (and maybe international) investigations. It currently only has a few members and is still in development – but I think it will become the next big thing in journalism.

A key feature which will aid its growth is the addition of a user profile. When you take part you give yourself tags to indicate your interests and skills, meaning you are easy to find when needed. Also, whenever you contribute to or start an investigation your profile is updated to reflect your increased participation. It is almost like collecting points or badges. This game like strategy will increase user engagement and provide a level of recognition that could be lost with other forms of crowdsourced collaborative investigation.

Help me Investigate has the potential to grow and as it develops similar sites will start to appear. The model of the website will get its first good test now that the government have begun the Coins data release. With such an abundance of government data ripe for investigation, Help Me Investigate can begin setting challenges to root out details from what has now been exposed.

Oli Conner blogs on technology and social media at SplinterNet


Who is bankrolling Britain's democracy? Which groups shape the stories we see in the press; which voices are silenced, and why? Sign up here to find out.


We encourage anyone to comment, please consult the oD commenting guidelines if you have any questions.
Audio available Bookmark Check Language Close Comments Download Facebook Link Email Newsletter Newsletter Play Print Share Twitter Youtube Search Instagram WhatsApp yourData