The what, the how and the money

Bari, 26 September 2013
Nicolas Kayser-Bril - @nicolaskb - Journalism++

Datajournalism is vast

Let's start with examples

What if

You received a hard drive
with 260 Gb of information?

From data to ministers stepping down: The offshore leaks stories

From data to politicians feeling uneasy

In a small Belgian town

How much do city mayors make?

Examples abound of data-driven investigations

France: An eye on MDs' income

Assessing the safety of Italian schools

Where did all those FDP voters go?

What do these stories have in common?

Some amount of data

Data analysis



Do you need a 3-people team to get started?

Anyone can start bootstrapping a data-driven story

Use off-the-shelf tools

TableauPublic, Datawrapper, Infogram

Up your skills!


Online platforms

Codecademy, Codeschool

Blogs, forums, mailing lists, books

Get talent from the outside!

Make friends with developers

Network with other journalists

The one skill you cannot outsource


These are harmless fun

but data illiteracy can cause great damage

Mistakes with risks: The UK pill scare

Biased data: the Orthodox Church

Statistical insignificance: e-cigs

How to get data literate?

Significant digits, stats basics, critical thinking go a long way.

How to get started?

Reuse the work of others!

Why work from the ground up each time?

Let free libraries do the heavy lifting

Taking datavisualization as an example, from most versatile to most simple

D3.js, Raphael, Highcharts, Chartbuilder

Or use off-the-shelf tools

TableauPublic, Datawrapper, Infogram

That holds for mapping...

Kartograph, Maps in D3.js, QGIS

for statistical analyis...

R, PSPP, Statwing

... for everything!

If you do it more than once, do a template.

Reuse from your own projects.

Some things that happen more than once.

Elections, storms, soccer world cups, floods, fires, olympic games, police investigations, cabinet reshuffles, Premier League tournaments, cabinet meetings, cricket world cups, trials, oil spills, Apple product launches Nobel prizes, murders, Champion's Leagues, riots, budget crises, food poisoning scandals, new car models, US bombing some country

Put your templates in one place.


And define metrics for success!

Structure your own data

Finding, refining and maintaining
a data set takes time

Organize your data sets & query them fast

Keep them in one place,
to be used by all the data team


SQL tables, OpenRefine, Google Docs

(Again, all free!)

Let's take an example

September 21st, 09:00 GMT

Westgate mall shooting starts

It takes 12 clicks to add context to the event

Because the cleaned terrorism database was at hand

Newsroom-specific tools

PANDA, LocalFocus

The database can become the product

Chicago Schools

Governement employees salaries

Connected China

How can one live from this?


Storytelling + developers

Journalism++ (Paris/Berlin, Stockholm, Amsterdam, Cologne), OpenDataCity (Berlin), Dataninja (Bologna), Effecinque (Genova), Dataveyes (Paris), AskMedia (Paris)

Corporate and advocacy storytelling

How many slaves work for you?


Read about OKCupid's blog on the O'Reilly Radar.
Read more on OKTrend.


OpenDataCity: Relog

OpenDataCity: Yosemite Fire


Hard to monetize

Traditional publishers set up "labs", but it's expensive

Summing up

1. Get started

2. Get data-literate

3. Leverage existing tools and networks

4. Be creative when selling


Your browser doesn't support the features required by this presentation, so you are presented with a simplified version of this presentation.

For the best experience please use the latest Chrome, Firefox or Safari browser.