#ONALondon keynote: What journalists can learn from game design

Speaker: Lindsay Grace – Associate Professor and Director of American University’s Game Lab and Studio, American University

Lindsay is a game maker and teaching games and interactive media for 12 years. He is an associate professor at American University and founding director of the American University Game Lab and Studio. More about his work here http://professorgrace.com/

He is interested in producing games that match the pace of news and engagement through play.

So he also kicked off his session with some facts

  • 1.2bn digital game players worldwide
  • 60-87% of the Western Civilisation have an active gaming community
  • It is a $100bn industry
  • Average player is 31
  • 48/52% male/female player ratio
  • People send about 60% of their time playing mobile games and about 4% on news but the reach is pretty much on par. They spend a lot of time, and a lot of money, on games. How to get them doing the same with the news?

He believes the driving force is engagement. There is a lot of content in the world (maybe not all of it good) but what makes content special? The answer: Experience.

Keeping people informed of the product of being repeatedly engaged; engagement design for news is informed by games design.

Content may be kind, experience is the kingdom. 

Game designers design experience, and games are useful for finding new experiences. They can inspire empathy (e.g. Darfur is Dying puts the player in the middle of the crisis), improve habits, develop skillsets, and help form the art of critical questioning. Game content is rarely as important as game experience.

Design the news experience and understand the many ways to engage the news community. For online dialogue many news organisations kick people out for ‘bad commenting’; we suggest you provide moderators with tools that are precise, actionable and effective, so they can carefully incise rather than smash.

Designing engaging news experiences. 

Consider ‘flow’ – in this state they lose track of time, but to achieve that you have to balance their anxiety and boredom states. This photo of the slide probably illustrates it better than any number of words could…



And what about monetising games? In the free to play mobile game space, minnows pay $0-$10; dolphins spend more, whales spend lots. Minnows make you popular, but whales make you money in the mobile gaming market.

In games there is something called Rubberbanding: Pulling 2 different players together for a shareable optimal experience.

Online and offline news standards and structures are relatively unchanged but his view is that you don’t just adapt the limitations and structures of print – you rubberband your audience. Think of ways to bring two different audiences (perhaps generational) together.

Employ dynamic audience adjustments so you use the real time reader feedback loop to make real time adjustments.

Engagement is dynamic: News is a dynamic system but we don’t always report it that way. There are 3 ways dynamics have been done in news – Interactives, toys and games – interactives provide feedback but aren’t playful; toys lack goals but can let you learn through play; games structure play by adding rules to how we do things. Game are interactive, playful and contain goals.

Games are really good at helping us make other people’s stories our stories – like news games around the Haiti earthquake, or being a Syrian refugee, but most of all games are about doing.

Among his case studies cited was Phone Story, where the player enforces slave labour, catches workers who kill themselves rather than work, and use hazardous materials that cause damage to people and the environment (Apple banned this gem because of its content).

Games help us understand the world in various ways. ‘Papers Please’ is a game where you play a border crossing guard –  the idea is that you are reviewing people’s lives to see who can enter a country. People think differently about immigration after playing this game.

Short experiences invite new people but long experiences appeal to the committed.  ‘This War of Mine’ depicts the civilian experience of war, and is a long experience compared to other games. Games and play experiences don’t have to be big experiences.

Angry Birds is something that people think of when we talk about games and successful games – Rovio made 51 bad games before they hit on Angry Birds. Games success formula is about failing quickly and failing often. Get it in front of people, see what they think, refine it on their feedback. You will learn more from your mistakes than your successes.

Games designers also respond to the force of the outside world – they will turn moves or books into games.



#ONALondon session: Reaching Unexpected Audiences With New Platforms


Erica Berger, who is founder of Catchpool and Co-Founder of Mileage Media, kicked off her session with some impressive facts:

  • 30 mins a day spent per day for average users of Snapchat
  • Instagram has 77% of audience outside the USA
  • Soundcloud has 350m users a month
  • WhatsApp has 990m users – 70% use it daily; it has 1m new users a day, 30bn messages spend a day
  • The average WhatsApp user spends 195 mins a week using it

Case studies 

Soundcloud: When Erica was working at The Economist she proposed using Soundcloud as a podcast hosting service, not least because it was free to use, and pushed it through to iTunes a s well – they garnered 827k followers within a few months and now the average 100k listens per track.

She started a similar initiative at The Week, which did modcasts – mini podcasts – and that now has 233k followers.

Catchpool: While at NewsCorp Erica realised the weekend editions were doing better performance than weekday ones. She looked at how newsletters curating quality links for leisurely weekend reads worked and from that success Catchpool was built.

NPR: How to attract younger users who would support with membership as well. The idea tested was how to get young people to listen together rather than in isolation, and discuss how the podcasts and shows made them feel. NPR went on a roadshow but instead of going to the north of America, they went south and found a really large audience. The Generation Listen initiative spawned an ongoing campaign that has led to new donations, and new supporters.

Her point? When you build tools for the people who are with you, you forget about supporting the people who want to be with you. Look for the unexpected audience.

The importance of measurement: 

1.Day 0:  Set goals

2. Day 15: Draw up a report

3. Day 30: Evaluate and assess if more time is needed

New platforms to factor into consideration: Blendle, Catchpool, Spotify, Soundcloud, Ryot, Outbrain’s chatbot

Erica ended her presentation with a short film discussing audience involvement and narrative shaping – watch it here – and a thought: Find the platform that is helping you get to those other places; Soundcloud can push your podcast to iTues, chatbots will work across several chat apps for you. You don’t have to do everything.

#ONALondon Session: Crowdsourcing as the ultimate strategic engagement

Panel: Mimi Onuoha, Research Fellow, Data & Society, Tow Society for Digital Journalism, Katerina Stravroula, freelance journalist and radio producer based in Athen and Tobias Dorfer

MO: Co-author of Tow report on the subject and before you read any further, you probably need to open this http://towcenter.org/research/guide-to-crowdsourcing/

Broadly defined by the Tow report as ‘The act of specifically inviting a group of people to participate in a reporting task’

People engaged in crow sourcing need to feel they have agency in contributing to a new story – we are not talking about scraping; people must not feel they are doing work for you. Everyone is getting something out of it.

The Two report says there are six types of crowdsourcing:

1. Voting – prioritising what reporters should tackle

2. Witnessing – sharing what you saw during a news event

3. Sharing personal experiences – tell us what you know

4. Tapping specialised expertise – contributing data or unique knowledge

5. Completing a task – volunteering time or skill to help create a news story

6. Engaging audiences – joining in call-outs (either informative or playful)

Crowdsourcing is about opportunities for communication via web technologies. As in the first session of the day it is about leveraging the collective intelligence of communities. People have something to gain, and it is it is crucial to the entire element of the entire story. It is just another part of the journalism process.

It is high touch, resource intensive and iterative. It allows journalists to tell stories that could not otherwise have been told, and it asserts the audience as an active participator in the story – the journalism is a relationship rather than a commodity.

TD: Engagement Editor at German news organisation Zeit Online which has about 10m unique users a month. User debates are at the heart of what it does and it has 67% of its audience aged under 49.

“Crowdsourcing is a really important part of our audience engagement strategy. It is not just about getting information that we might not be able to get otherwise, but it is about credibility. Crowdsourcing is the possibility of giving our readers the chance to be a part of project rather than consuming a product and it allows us to gain trust”.

Case study: Zeit Online investigated overdraft rates and asked the audience to send in their postcodes, BIC and overdraft rates. The project got 10k participants, information about 691 banks, and from that created a map of the two highest overdraft interest rats for every state, alongside several articles.

Case study: Bakeries that bake bread on the premises 

Crowdsourced for readers to share the bakeries they knew of that sold handmade-on-site bread and rolls, and what baked goods they recommended. 15K participants, with more than 2,500 such bakeries identified. A map was created and included readers’ favourite products from those bakeries.

With these two case studies, Zeit Online used Google Forms and ran the campaign call outs for around two weeks.

KS is a member of the radiobubble.gr community, contributing to the rbnews and rbdata teams. Radiobubble comprises journalists and activist in Greece.

She was part of the team behind Generation E, a data driven investigation into migration; with a small team she crowdsourced for people to tell their experiences and stories. (The E stands for Europe, Emigration, Erasmus, Economy, Exodus, Escape.)

The team wanted official data, but also the stories of those impacted by it. We used a form online – we first invited people to participate, and tell their stories alongside sharing their data.

They received 2,400 stories and their top level findings included the driving factors for emigration, the registration of non-European migrants, and the inclination emigrants had to return to their home countries.

Tools: Open Refine, Datawrapper, Trello, GoogleDrive and Forms. Doodle, Twitter, Facebook, and also the team worked with media partners.

Takeaways from the panel

You need to plan but be prepared for what comes back to be different to what you expected

If you have a data journalism project, as a freelancer, you cannot continue indefinitely without funding



So, a few thoughts from me as a result of sitting through this very rewarding session…

Personally I think the big point journalists can take away from this was just how much information and time people are prepared to share – IF you can hit a topic they care about.

A few years ago, when I was editing the Daily Post, we ran a survey on a notorious road, the A55, but – crucially, I think – as well as asking people if it should be given a 3rd lane (no brainer answer: Yes) we asked them to share their views on what the biggest problems were, what they thought should be done, and any experiences they wanted to share.

It was incredibly successful in terms of response and richness of detail and made for several days with of content.

Crowdsourcing, I think, means you cede control of your questions and your line of investigation – what people want to tell you about may be only indirectly linked to the question you ask initially, but if you follow that line of inquiry, you may find the rewards, engagement and validity of the journalism is far, far greater than you imagined at the start.

Crowdsourcing is not “send us pictures of your children in Easter bonnets”; that’s a UGC shoutout. It is the collaborative act of putting inquiries into the world, and seeing what develops – of making stories with people who are outside of your newsroom and your bubble of perception.

#ONALondon Session: It just got personal – engagement via chat apps 

Speaker: Sarah Marshall, Social Media Editor, The Wall Street Journal, EMEA.

I’m going to make the bold prediction – with the afternoon still to go – that this is my favourite session. I learned so much, and enjoyed the speakers enormously although sadly I didn’t catch the name of the Globe&Mail ad hoc speaker.
Sarah (@SarahMarshall) gave insights into the WSJ’s use of Snapchat; it lunched on Snapchat Discover on January 6 2016.

Why are the WSJ doing it? 1) New and younger audiences; 2) revenue stream (the WSJ and Snapchat sell vertical ads; research shows most people also do not find ads in Snapchat annoying.)

Snapchat has 100m daily active users, of which 60% aged 18-34 and 44% are using either Discover or Live.

We publish 5 days a week on Snapchat but lots of our audience want us at the wekend. 8-10 stories per edition and a team of 5 people roughly, but they work across teams and about a third of my job is creating news for Snapchat.

The Journal publishes all sorts of content but to distinguish ourselves we concentrate on Business, Markets and Tech. We do some sport and world news but those are our 3 verticals. WE are contractually obliged to provide 5 original stories a week.

On a Friday we make an edition where the one edition tells the whole story.


How do you have a flow to the narrative so people want to swipe through? Having regular features helps the audience form a habit, and come back and know what to expect.


We have a much younger audience. We write for the platform not the audience but we know who that audience is and think about it as maybe they are at an Ivy League university or starting their first job at Goldman Sachs. We are after people who will become WSJ readers, who are ambitious. We will always have smaller audiences than, for example, Cosmo, but we will have – we hope – future subscribers.

The three key words I think every day in relation to our audiences – inspirational, aspirational, entrepreneurial.

Our audience numbers are huge and the loyalty is amazing (Sarah isn’t allowed to share the figures but stresses that this happened more or less overnight. A new loyal audience that came every day was created from a standing start). This is the biggest thing the Journal has done in terms of audience development for years.

Unexpected outcomes

How sometimes we can right stories in a much tighter way. I quite often take a 1,000 word story on corruption in Malaysia and condensing it down to key components which is maybe 300 words. Most stories can be told really well in 300 words, and the learnings of that will go out to other areas of the newsroom.

How we work across other teams – the workflows have changed. I sit opposite the people who work on Whats News and if I condense a story into 300 words that goes to them, and vice versa. When you allow teams to work together it is amazing what can happen.

There are finite places on Discover and not every news organisation can join. We started on Snapchat Stories (as TeamWSJ)

We try and ask questions we really want the answers to, or really care about.

Before we launched we asked journalists to think about pitching stories for Snapchat and then we launched… and suddenly it was really easy because everyone could see how it would work. Adding bylines also made a difference to how much people wanted to get involved.

Growing an audience is hard because it’s not shareable like Facebook, and we’ve been helped by Discover. However, you see other titles doing great stuff every day – 30% of people surveyed got their information on US elections via Snapchat. It is a slower build but it is worth the time and effort of thinking how to do it for your particular brand.

3 Takeaways from Sarah:

  1. Consider the voice for the platform
  2. Enable creative workflows
  3. Play the long game

Case study for WhatsApp:

The Globe and Mail in Toronto ran a WhatsApp campaign around the elections. Readers responded well and gave good feedback asking for more. They found it a convenient way of getting manageable information. However, it was difficult to manage the workflow. (WhatsApp limits the number of people you can add to a broadcast list). It was admin heavy but for the G&M but when WhatsApp do make it more effective for larger scale us, it will already have a leg up on competitors.

Case study for YikYak (from @james_morgan)

The BBC is using YikYak, which is extremely well established in the USA and Canada. It is anonymous and also local.

BBC thought that as it was a 98% millennial audience it would partner with YikYak and in Canada, for the elections, use the Herds feature to start a conversation around the topic.

The BBC was worried about how it would play out, but actually got tens of thousands of responses and used the best on its live blogs. More recently, with Newsbeat brands, the anonymity feature was used to discuss mental health and again it proved successful.

The BBC was concerned about verification because it was an anonymous platform, but the community is highly self-policing using the up/downvote options.

It is no longer anonymous and users can identify themselves – whether this will be a plus or kill what makes YikYak unique remains to be seen.

#ONALondon ‘When news breaks bad: UGC in the newsroom’ session 

Panel: Mandy Jenkins, VP ONA board, head of news, Storyful; Fergus Bell, ONA Ethics committee, and Dig Deeper Media 

Ok, so for this blog post to make much sense you probably need to head straight to http://toolkit.journalists.org and have a look at that, because this is what we’re talking about.

Done that? Good – the roundup from the session starts… now:

In 2012 the first discussions around UGC and eyewitness protection began.

FB we didn’t have the luxury of several years to develop; we needed to come to a level of understanding quite quickly, and so we set up a working group and many conversations have ben held over theyears.

MJ: We are constantly dealing in a world of unknowns and what we have learned has meant we’ve started to figure out the day to day world of news.

The ONA Social Newsgathering Ethics Code is a document to gather the support of news and journalism organisations internationally to endorse a set of standards and practices.

Here’s a quick screen grab:

ONA Social Newsgathering Ethics Code – ONA s Digital ToolkitExample: ‘The Eiffel Tower has gone dark’ – many news organisations were posting on social in the aftermath of the Paris attacks that the tower was switched off out of respect. Actually, the lights are switched off every night.

Example: The video bandied around as being of the Brussels terror attack that turned out to be Moscow, 2011.

MJ: Readers and journalists are coming to standard terms that indicate where we are in a breaking news story. We will say ‘confirmed’ and how we know what we know, or ‘unconfirmed’ or ‘checking’ which says ‘we are looking into this right now’. The transparency is there and our audience can see what our status is with regards to verification of a story.

FB: For journalists, thinking of how we say we are reaching conclusions around stories is an alien concept but what we need to get our heads around is that the audience is now searching social media themselves, and if it doesn’t look as though you are going through a verification process as a professional journalist they may well question why.

Considering the emotional state and safety of contributors is about the way we deal with people who are creating content we need, and who we are using to tell stories.

FB used the example of a campus shooting eyewitness who was asked by hundreds, if not thousands, of media for her experiences – while it was still going on.

Journalists were getting abuse from other people who could see them asking for content.

Storyful approaches it like this:


FB: Be aware there is a difference between getting that story when someone is in fear of their life, and when they are safe. If you are communicating with someone who is hiding from a gunman, as in this case, what happens if their phone is not on mute and it makes a noise when they receive a notification?

Or if there is a particular geographic reference point or angle on a photograph that shows where they are?

MJ within your own newsroom there needs to be communication to say if someone has reached out to an eyewitness, so they are not getting bombarded.

Assignments v discovery – asking people to create content for you is different to finding content they have made. A snowy day photo shout out is low risk – a hurricane pic shout out is not.

MJ: during the Kenyan mall attack you could see people hiding while journalists reached out to them asking them to shoot video. These people were being asked to put their lives at risk to film for a news organisation. It was very ethically unsound.

Storyful works with content that exists and does not create assignment situations (i.e. asking for content to be created rather than provided post-creation)

FB: I worked with UGC on various uprisings in Libya and other areas and we would never ask them to capture anything that was live. What they captured informed the reporting rather than vice versa. We had a responsibility to their safety.

It is not only about protecting the source who provides the content but also protecting those who are featured in it – like making sure witnesses captured in a video are not identifiable.

Embedding can also be an area that needs negotiation: Sometimes people also don’t realise their content is public – they think it is contained within their network and don’t expect to see if in other media.

In breaking news if you are asking for permission to use something, do you have archive rights? Use beyond one time? Multi-platform? What if people later change their minds? If someone retracts at any point, if you’ve negotiated use on Twitter, you have to comply.  (This is a key point that newsrooms need to understand, as far as I am concerned)

If you throw a bunch of legalese at someone around usage, you aren’t likely to get anywHere.

By being more ethical you can also be more effective.

Keeping journalists safe online

MJ: I have had a member of staff stalked by someone they reached out to in pursuit of a story. We have to know that we are not supermen and superwomen when it comes to dealing with this.

FB: I know investigative journalists who will meet some potentially unsavoury contacts in public places with colleagues nearby. If you have a junior staffer reaching out through social to people, as newsroom managers do you know that staffer’s exit route? Do they know how to protect themselves?

Not everyone is bad out there of course but perhaps if you are able to speak to a source on the phone or company email rather than expose your personal account is useful in some cases.

*Also see the work by the excellent Eyewitness Media Hub. I’ve been some small involvement with this, and and blogged about it here and here.

#ONALondon First session: ‘People you may know: Your audience’

I am at ONA London today, where the theme is audience engagement. Rather than live tweeting it, because there is SO MUCH to share, I thought I’d take notes and post them here.

First Keynote session: People you may know: Your audience 

Panel: Federicca Cherubini, Digital News Project team at @risj_oxford (among her many other hats) Mary Hamilton, executive editor, audience, with The Guardian, Renée Kaplan, FT, head of audience engagement 

The point of this is what do we know about our audience, how do we know it and what action do we take as a result of that knowledge.

What does it mean when we say we want to harness the power of the audience?

RC  it has become such a slick term that  it begins to mean nothing. For her it is the idea of a relationship and thinking of distribution as a relationship. We know the audience is there but there is now a notion that we also have to be there giving them something that they value. Putting a metric to each part of that relationship so people can see how they are performing against value/delight/quality of relationships.

MH working out what we want people to do in relation to what we produce; what is the target and then how we engage in a 2-day conversation. It is different for every news organisation. Understanding the ‘what do we want people to do in response to what we are doing?’ is fundamental.

We look at reach, optimisation, loyalty, engagement, habit, participation.

RC internal engagement is also a factor – the culture change and the different ways of thinking are important within this definition too. She also spoke about Lantern, the FT’s new tool which assists journalists in accessing data about content and interaction.

What is loyalty? MH says it varies per person – it might be a subscriber, a regular user or even a sentiment. As the Guardian moves more towards membership it is looking into what is the loyalty behaviour it would expect. RC says loyalty begins with the question of what do you want your audience to do? Loyalty for us means that eventually someone gets enough value from our content that they will subscribe. She also raised the very valid point of the newsroom needing to understanding the commercial importance of audience interaction.

MH says page views are still very important, not least because they are very easily understood across newsrooms (I am so glad she got that point across because they 100% are). But it is easy to talk about the big number and then ignore the 2nd step -do they come back, do they click on something else?

She also had my favourite phrase of the day “lumps of behaviour” – it made perfect sense but also inexplicable made me happy.

Reach is important but depth of reach is as important.

Comments: RK – comments are very important to us because the people who leave comments are our subscribers and are invested in us.

The people who comment on stories are often the people who care most, and that might be that they are very angry with you because you haven’t done what they would expect of you. You have to be present and listen to the comments and take those words into account.

However, comment moderation doesn’t scale in the same way – if you are going to do it well you need skilled, thoughtful moderators.

Update: You can also read Sarah Marshall’s notes here


So, my takeaway from listening to this (really excellent) session: Listening to these very smart speakers, it struck me that if we referred to audience engagement within our newsrooms as audience relationships it might be a concept that is more easily grasped as existing beyond metrics.

Relationship is a term that we farmed off to the marketing years ago (the CRM) and have only recently (in terms of digital journalism) clawed back.

For a vast chunk of my career my relationship was definitely internal (writing for my newspaper, my news editor, my peers, my sales department). How insular was that?

Audience engagement is a lot of work, when you think about it.

You have to scrutinise metrics and listen to the audience to see what they care about, respond with crowdsroucing and a piece of content, work out the best places to put that content so people see it and ideally share it, interact once that content is published, adapt it to reflect what the audience is saying about it – maybe also stand up for what you have created or change the view that you initially approached it with – and after all that, if you’re anything like me, no matter how many people say something is good you will brood on the one person who says you are disgrace to your profession.

It’s a real tough sell to people in the newsroom who are not really invested in the strategy/concept/demands of audience engagement or who are just not used to getting feedback beyond the praise of the person sat at the next desk.

More and more I think the best investment in training a news organisation can make is to help people wrap their heads around their audience strategy, and how individuals (both those working for titles and the audiences responding) have a role to play within that.

Oh, yeah – and there also needs to be utterly clear leadership from the top of the newsroom down around the importance of it. If the top bosses aren’t keen on getting to know and being involved with the audience, of course the wider team is going to take a cue from that.