5 lessons on building a newsroom from scratch

In the early days at Storyful there was a mission to build a team (and later technology) to monitor the globe for breaking news events. The mission could be summed up as:

…find all breaking news events likely to generate eyewitness content (videos particularly), in any language, in any geography, at any time, from any device, on any platform, with the smallest number of journalists possible – and verify that the content is real, and seek permission to use it. Do this without access to any traditional wire services and solely rely on social media and free tools to detect, source, verify and clear (and later licence) content.

From an office in Dublin with a small team (less than 10 in 2011), that’s quite an ambitious task. But we were largely successful at it. There are a number of things you look back on at a startup and say: we got x right, but we got y wrong.

For this I think we managed to more or less nail it – along with an office culture that re-enforced each of the points I go through below. I consider myself lucky to have the opportunity to build something like this from scratch – it was an incredibly rewarding experience.

I worked at a newspaper for three years before joining Storyful – or as it was then, an as-yet unnamed startup company. But I came from a blogging culture, having started a self-hosted blog way back in 2002. It’s an interesting contrast.

At a newspaper (the Irish Examiner) I would often puzzle at why certain things were done the way they were in newspaper production – and sometimes found the answers odd or amusing. “Because that’s the way we do it” was one. “Why are you trying to change things?” was another.

There were a few funny moments – I recall once suggesting that we tweet an image of the page one of the next day’s newspaper (this was 2008) – a tease to our readers. I received a look of almost friendly contempt for it from my then boss (they later started doing it in 2011). There was generally a lack if willingness to experiment – which often meant I ended up experimenting myself. (There was also something of a Twitter rule among the hacks: The hacks who most resisted or most mocked journalists using Twitter, were the most stalwart converts once they figured out how to use it.)

But being a blogger – particularly a self-hosted one on gavinsblog.com – I was forced to learn how to do things, and how to adapt to changing circumstances, and always to try new ways of doing things. Self-hosting with vanilla HTML, then Radio Userland, Movable Type and then WordPress forced me to consider various technologies and approaches, and to get your hands dirty in SQL databases, PHP, HTML or CSS.

You’re confronted with questions: How does linking work? Why did this headline get me so much traffic? How do I build a community of loyal readers? What does my brand mean to my audience? Should I use ads?

I would have learned much less had I just had a Blogspot blog. But I was forced to learn more by trying and failing.

So it was with a mix of these different philosophies and experiences that we started to build the Storyful newsroom. One of Ireland’s first truly 24/7 newsrooms – and one that was entirely digital – it had no newspaper output, no TV output – but was a pure agency model. I think the relative success of the newsroom could be summed up under the following headlines, but is by no means comprehensive. (I no longer work there either, so some of these have evolved!)

First principles

When you have limited resources and limited time, it helps to try what could be described as working from ‘first principles’. This aims to boil down your objectives to the simplest ones possible.

Having studied philosophy before anything else in my life, I think of this in the Greek/Aristotelian philosophy sense (rather than the physics sense Elon Musk is a fan of). When you boiled down the task of our newsroom, the job was:

  1. To source information
  2. To filter and parse that information
  3. To distribute the results of 1 and 2.

These could be called our first principles. It is also how full stack development teams were organised after the acquisition (though this has evolved since).


The second major view was on transparency. I brought a certain amount of baggage with me on this issue, as I was at the time becoming quite an FOI nerd. But this philosophy was reflected in both our constrained resources and a deeper view on admitting that journalism is by its nature imperfect and incomplete – and almost always stays that way.

This led to what perhaps now could be described as the Storyful news agency style: “Here’s what we know; here’s how we came to that conclusion; here’s what we don’t know; here’s what we’re trying to establish.” Or perhaps more succinctly in the early days: here’s stuff we know or can back up; here’s what we don’t know (yet).

In terms of our early style of disseminating information to our clients we took a particular view about our limitations – be honest about them, explain them, explain what we’re working on – but show our work by saying what we have established and how we did it.

Transparent by default.

(I also had ideas that we should not just be transparent, but radically so – that our output should default to versioning, so our clients could see the earliest drafts of our output, all the way up to the current and evolving version (not unlike Wikipedia). Even to see when we’re typing in realtime, not unlike ICQ in the early days. We never got round to implementing it.)

I think these principles could applied to most newsrooms – just in differing ways depending on the output.

Editorial Team

My experience in newspaper production meant I had worked with some really good copy editors (I was never that good). My experience in blogging meant I knew some good bloggers too. The mix of the two – or at least copy editors unafraid of technology – turned out to be the ideal candidates for hire. Many early hires were mostly from my existing network.

While Mark Little‘s job as founder/CEO was partly to set the vision for the team (he’s really good at it) and particularly in the early days to go out and sell the service, the job of the editorial team back in Dublin was to execute on the vision. That fell mainly to me towards the end of 2010 (along with Mark Coughlan‘s stint at the company), and Markham who we took on in Autumn 2010. We started scaling the team rapidly in early 2011, amid the start of the Arab Spring. (I should preface by saying that some people came and went – some really good people and some people who didn’t work out for various reasons. The list below is the people I was personally already connected to, or were connected to through my own existing network).

  • Markham was a blogger and former copy editor and I knew him through blogging. Perfect. He later became Managing Editor and left around the time of acquisition – he’s now head of Visual Storytelling at Vocativ in NYC.
  • Eoghan was a former colleague at the Examiner – and I knew him to be a constant experimenter with technology (we both loved Flickr in the early days). He still works at Storyful.
  • Felim was my other former colleague at the newspaper, who had a penchant for detail, and while not a blogger, he was a constant tinkerer with technology (he now works at the UN).
  • Malachy was someone who I had communicated with a few times about various things – but we had similar career trajectories between tech and online publications. We went for a pint and got on right away. He later became News Editor, and then to Reported.ly as Europe Editor and now works at The New York Times. 
  • The extended network kicked in then – with Alan joining from the International Herald Tribune because he was a known quantity to both Eoghan and Felim, with whom he had worked in a different company before. “Is he good?” I asked the lads. “Excellent” they said. Sold. (Alan is now standards editor at Storyful).  Joe – now Director of News at Storyful – had previously interned for Malachy. “He’s brilliant,” said Mal.
  • Aine was more on Mark’s network than mine – although Ireland being the small place it is, me, Eoghan and Felim had all edited her copy when she was a reporter at the Irish Examiner. (Aine started mainly on US politics, later started the viral team, and then became Managing Editor. Aine is now manager of news partnerships at Facebook.)

All of these people were the right mix of technical skill, ability to learn and eye for detail that the job required. They were also all open to the demands that a startup requires – extra hours, less than ideal working environment, and happy to take a risk on a company with little revenue.

Control of technology

I recall that Storyful had a printer that the Mark brought from his house. Printers were not something we used much in the office, and it kept breaking down – and its use was mainly for business stuff. For the newsroom though, printers were an anachronism. Why would we need a printer, when we could just use Google Docs for everything?

Control of technology is extremely important to allow the newsroom to adapt new workflows and try new tools. In many newsrooms I’ve visited or worked in, the “IT guy” tells you what you can and can’t install on your desktop, or if you do want to install something, you have to go through a bureaucratic mess that takes weeks. This has an incredibly negative affect on experimentation – and while it might be often for good reasons, security being chief among them – it is often applied too robustly, or the IT guy lacks the empathy for what the editorial team needs.

We had no such problem, in fact I think Storyful never actually hired someone to look after the IT of the company until it was over 5 years old (after we were acquired by News Corp). I configured and bought most of Storyful’s early machines (I think they’re mostly out of action by now). We needed fast machines (so Core i7s), Windows not Mac (we were a startup after all), ideally 8Gb of RAM, support for at least two, if not three screens natively – with high end graphics cards. Along with plain vanilla Windows 7, Google Apps for Business, and a bunch of free tools we standardised for every member of the team.

In general nothing needed to be stored on the machines themselves, so whenever we had a problem with a machine, I would simply wipe it and clean install Windows. This has the added benefit of removing any possible malware or spyware. We were able to keep at first a half dozen – and later more than a dozen machines – running reasonably well – for years. The introduction of things like shared Chrome profiles meant that we could standardise extensions and bookmarks across the newsroom too, without having to rely on local backups.

Having control of your own technology – and knowing how to use it effectively are obviously important things for a newsroom. We were able to control our own tech safely because we were nerds.

Unfortunately I still hear of newsrooms – famous and big ones – who haven’t realised that controlling your tech is really important. (Perhaps they still see computers as connected typewriters? 🙂 )


One of the core beliefs and philosophies that I brought with me from my time at the Irish Examiner was that there was inequality of skills in newsrooms. Some journalists knew how to do certain things really well, other journalists – in the same room as them – had no idea how to do precisely the same thing.

This is enormously frustrating to witness. And it’s equally frustrating if you’re someone who is eager to learn. It’s even more frustrating when certain journalists actively refuse to share skills precisely because they don’t want their colleagues to be as skilled as them – lest they have competitors in their own newsroom.

When we were building the newsroom at Storyful we had a relatively formal rule, along the lines of:

Everyone in the newsroom must have the same skill level as everyone else. No one should have any deficit of skill compared to any other colleague. When a new team member joins – regardless of being an intern or new hire – they will be expected to learn every skill at the level of the most senior editorial member. If any member of the team feels lacking, the onus is on them first to ask, and then on the team collectively to ensure that person is brought up to speed.

I was pleased to listen recently to a Storyful podcast, and hear from interns and staff that this philosophy is still at the core of how the newsroom functions today. Partly this rule was also out of necessity – in such a small team, everyone had to be at the same level. But it was also that philosophically – in my view – all members of an editorial team should have the skills necessary to bring to bear, should it be needed.

We took a lot of care in the early days that when new people joined, a good deal of hand holding and shadowing occurred, often for weeks – to ensure that the most obvious questions were answered – that new hires knew that it was safe to ask questions and that there was no such thing as a stupid question.

It was also made clear by inference that all of us were on a learning curve in a rapidly changing industry – and that by having the foundation of all being on the same skill level we also implicitly all had valid views on new or better ways to work. No one has a monopoly on innovation (notwithstanding my cheesy title of Innovation Director).

If modern newsrooms are to succeed, there must be a willingness to share and learn – regardless of seniority, how long you’ve worked there, where you worked before, or how good you think you are.

I think we managed to pull these five aspects together well (along with many others), all while the industry was changing rapidly – and we were always agile enough to keep changing as the business evolved.