Did you hear the news? Five innovative ways of implementing audio in newsrooms
The second half of 2017 has been saturated with talk about news organisations investing significantly in video. In all that talk and speculation, I noticed an important topic being overlooked: audio. Here, I’ve looked into some recent experiments in digital audio news and podcasting I was curious to learn more about.
I spoke with João Pedro Pereira and Diogo Queiroz Andrade from Público (Portugal), Taseer Afzal from Dinbar (India), Henry Cooke from the BBC, and Mona Chalabi and Josie Holtzman from the Guardian US.
This piece is a rundown of some experiments in news and audio that caught our attention. It’s not a definitive list of trends to watch.
Not all these experiments are on the same scale or will have the same impact. Some have bigger industry-wide implications, some are just projects I’d be curious to see more folks experimenting with.
1. Short daily catch-ups
Daily audio catch-ups, whether distributed natively or within third-party apps, are gaining ground.
Público, a Portuguese legacy media organisation, is currently beta testing P24, a personalised audio catch-up format. The project is Público’s first large-scale personalisation feature and part of a larger-scale ‘pivot to audio’ for the brand.
Why audio? P24 aims to deepen their relationship with their existing users who were only spending a few minutes per day on Público’s site.
‘This product was made to serve the kind of users we already have, but we wanted to serve them better’, said Pereira, Público’s innovation coordinator. P24 creates personalised daily digest packages that are less than ten minutes long and ‘carry our journalism into moments of people’s lives which are not typical moments that they would be with us. We can be with people when they’re at the gym, in their cars, or washing their dishes’.
The personalisation feature of P24 is native to Público’s website and requires users to log in and choose the topics that interest them. Journalism.co.uk reported that ‘P24 uses a mix of algorithmic curation and editorial input, as a team of Público journalists headed by an editor make decisions on which audio pieces are selected, to ensure important news gets included even if a person has not previously expressed an interest in a particular topic.’ P24 is also offered as a daily podcast, though without personalised content.
P24 is 70 percent funded through Google DNI and its timeline includes a full year of beta testing from April 2017 to 2018. Worth noting, BMW has been a P24 sponsor since its launch and ‘wanted to join precisely because of the innovation factor’, not big audience numbers, said Andrade, Público’s deputy editor Diogo Queiroz Andrade.
P24 is part of a larger audio pivot for Público, which includes investing in podcasting. Público now has a total of eight podcasts with three more set to launch in 2018. They’ve set up an audio team of eight: seven journalists (mostly young unknown talent) and an editor.
Público hopes to organise the country’s first podcast festival soon.
‘We believe that audio interfaces are growing, so we are betting on them to become very relevant. Not in Portugal yet — mostly in Anglo-Saxon markets — but we know they will grow here too. We want P24 and our whole audio offering to be sustainable. We want to be a reference in terms of audio in Portuguese news and we want P24 to lead that’, said Andrade.
The Skimm is another media brand that’s betting on audio. Launched in 2012 as a daily newsletter, The Skimm recently introduced Skimm Notes: ten-minute, topic-based audio explainers that are available only within their native (and paid) app. Like Público, The Skimm sees the potential of audio to connect with users where screens fail.
‘Eighty-six percent of our audience is commuting regularly. Eighty-two percent listen [to some sort of content] on their morning commute and 74 percent listen on their evening commute. When we thought about that, we knew we had to step into the routine’, The Skimm founders told Nieman Lab.
2. Distributing content via closed chat apps
Dinbar is a short (around seven minutes) news bulletin in Urdu distributed daily and exclusively via WhatsApp. It was the brainchild of news junkie Taseer Afzal, who launched the bulletin with two friends back in 2014. To make the bulletin each day, Afzal’s two friends do the reporting legwork and he records only using a smartphone app. Afzal said in an interview with FactorDaily:
‘In this age of smartphones, people don’t have the time and patience to listen to news on the TV or radio. Our Whatsapp audio news bulletin can be downloaded any time as per the listener’s convenience.’
Dinbar, delivered through dozens of WhatsApp groups, reaches around 12,500 users – residents of the Indian state Jammu and Kashmir as well as displaced Kashmiris in India and beyond. The groups are run by the three founders’ friends and contain hundreds of members each. The ‘dark social’ nature of these closed groups make it difficult to tell how many additional shares each bulletin gets but Afzal estimates that it could be as many as 200,000.
Dinbar is free for now, but the team is considering a small subscription fee in the future which could allow a bulletin in Kashmiri as well as Urdu. But when it comes to selling ads on the bulletin, Afzal is not interested.
‘I don’t want to because it will be longer’, he told me over the phone, ‘Some areas are without 4G [and] we want it to be as light as possible. Six to seven minutes is 1.5 MB, that’s why downloading is easy’.
That’s another advantage of distributing audio: data-light content will always be more accessible than video and cheaper to produce.
3. Smart speakers for interactive audio
Last month, the BBC launched The Inspection Chamber, an interactive audio sci-fi comedy exclusively built for Amazon’s Alexa. Listeners use their own voice to become a character in the story and influence its outcome.,
‘Our hunch was that direct conversation that changed the details of a story would be more satisfying than making choose-your-own-adventure style choices. Playing a part, rather than directing the action’, Henry Cooke, senior producer for BBC R&D, told me via email.
In a September 2017 blog post on the BBC, Cooke also wrote: ‘We wanted to make it feel like you’re having a genuine, direct interaction with the other characters in the piece. We haven’t come across any other interactive stories like this on voice devices, and we’re excited to see how people respond to it.’
The BBC, in collaboration with private production company Rosina Sound, built The Inspection Chamber within the constraints of this new medium. For example, the story was scripted in short segments with frequent questions because, according to Cooke, Alexa and other voice devices aren't built for long-running conversations as entertainment. They also used story serving techniques borrowed from video games to keep track of all the potential directions the narrative can take.
Right now, the team behind The Inspection Chamber is collecting feedback via BBC Taster.
‘User testing and analytics from the live skill are showing us how people respond to interacting with stories on voice devices, and tastes in things like types of question and lengths of listening sessions’, Cooke said.
In the future, Cookie said the BBC will examine the types of content that works well on smart speakers and accordingly create new experiences that are tailored for these platforms.
I asked about the potential for this format for news. Cooke was optimistic.
‘It’s easy to imagine how you could interact with a news piece to ask for greater context, more detail and related items when you feel you need a bit of a bigger picture about a story’.
4. Liberating audio from platforms
Another audio trend to watch is news organisations breaking podcasts out of dedicated platforms and apps. Currently, you have to download an app in order to subscribe to and download podcast episodes, but this breakaway makes for a more streamlined user experience
‘It’s been fascinating to see how easily you can build an audience on the web when you remove the barrier of having to download and use an app’, wrote Alastair Coote from the Guardian Mobile Innovation Lab.
Coote used an in-house hack day to challenge himself to improve podcasting UX through a web-based mobile player on the Guardian’s site. Coote combined the Media Session API (to integrate a media player into Chrome) and created a subscription option via push notification to notify listeners when new episodes are available.
‘Zero friction, and hopefully far less confusing to users who don’t normally listen to podcasts’, wrote Coote.
While this project visibly (audibly?) remains in testing stage, other publishers are jumping on this trend.
The breakaway from dedicated podcast consumption and hosting platforms will also give publishers the ability to track ‘listenership’ data. For instance, Apple’s listening metrics are notoriously opaque. This means that ad sales must be made based on the number of downloads – and not much else. (To remedy this, Apple announced new granular in-episode analytics this past spring, but so far rollout details have been fuzzy.)
It is because of this that media brands have been trying to bring listeners (along with their data) back to their own platforms. Chiefly among those is The New York Times, which announced yesterday that The Daily and other Times podcasts are now natively available in their app.
In a press release, Associate Product Manager of Multimedia Jordan Vita said that ‘having a native player opens the door to new opportunities to improve the audio listening experience for our audience’.
NPR has also been experimenting with a project they’ve aptly named ‘Remote Audio Data’ on their app NPR One. It allows for more fine-tuned listener data without relying on third parties. Producers can, for example, know where listeners dropped off or whether listeners heard or skipped ads.
‘We don’t have to settle for knowing the magazine made it to the mailbox’, wrote National Public Media COO Bryan Moffett.
Let these brands lead by example. As quoted in Hot Pod, Edison Research’s Tom Webster wrote that ‘the industry cannot and must not stop innovating towards a non-platform-specific measure’.
5. Data storytelling for audio
When it comes to data storytelling with audio, good examples are unfortunately few and far between. Data sonification (the use of non-speech audio to` convey information or perceptualise data — thank you Wikipedia) for news has a lot of untapped potential. Audio is linear and thus lends itself well to timeline-based data as earthquake frequency in Oklahoma or the scramble towards a deadline for a research project. BBC was publishing short data pieces they called Audiographs back in 2016 but the project seems to have been dormant for the past year. Some pieces of data sonification like Climate Symphony or Listen to Wikipedia are interesting but seem more artsy than informative.
Luckily, Mona Chalabi is coming to the rescue. The notable data democratiser is trying her hand at audio in collaboration with multimedia producer Josie Holtzman. The pair are launching a new Guardian podcast early next year.
‘The starting point was that we wanted to make a podcast that communicated numbers in a new way,’ Chalabi told me.
Chalabi is known for her very visually intuitive data sketches, often based on reader inquiries. The upcoming Guardian podcast will be in a similar style, taking cues from her readers of ‘trying to find a number and find out whether or not that number matters’. The episodes will rely mainly on narrative storytelling rather than data sonification.
‘One of the things that data can’t do sometimes is really capture the full human experience of the numbers. The impact that those numbers have. What we really want to do is to make a podcast that takes a number or takes a set of statistics and explores them from a very human lens,’ she said.
Chalabi sees audio as a ‘really exciting challenge’ and can apply the best practices from her visual data journalism to the medium. She underlines the importance of editing, choosing the most important data points and making sure they’re communicated to the user effectively.
‘If finish a chart and I annotate every single bar on the chart with a number you are going to lose interest in that chart very quickly. And same with audio. I can’t narrate every single number to you. Figuring out ways to do that is really interesting.’
Sometimes visualisations lack the nuance required to tackle certain subjects‘Let’s say someone wanted me to do a chart on gun violence. I don’t want to do it with chalk outlines or something on that visual chart. I think audio allows you get some of that sensitivity by simply allowing people to speak for themselves and asking them the right questions. That to me is really exciting. Not just putting out statistics, but to explain the impact the people that made that number what it is.’
Don’t count data sonification out of the project completely. It’s a feature both Chalabi and Holtzman are keen to incorporate.
‘Part of the goal was to initially create a chart that works visually without the audio and works with audio without the visual’.
As with her visual style, she want any audio data to represent the given subject ‘not to be like dings’. They’re planning to use the Guardian’s web-based player (Coote’s, in fact) to serve visuals at specific moments during the playback. The marriage of audio and visuals will also help with shareability, a tricky problem with podcasts.
‘Some of it is about teaching you to expect something new’, said Chalabi. ‘Both of us will be focused on something that’s sharable.’
Our ears await.
Diogo Q Andrade is the Deputy Editor of Público.
João Pedro Pereira is the Innovation Coordinator at Público.
Henry Cooke is a Senior Producer & Creative Technologist in BBC R&D, where he leads the Talking with Machines project.
Taseer Afzal is the founder and host of Dinbar.
Mona Chalabi is data editor at Guardian US.
Josie Holtzman is a Brooklyn-based multimedia storyteller and freelance audio producer.
This article was updated at 13:00 (CET) 15 December 2017 to include audience data from Público.
