Hi Alexa, is the monetisation conversation moot?
Voice assistants are still young and a bit awkward. Content experiences on them aren’t always great and it seems that news organisations, marketers, and the general public haven’t quite figured them out just yet.
But voice AI technologies are on their way to massively changing access to content. ComScore has gone as far as to predict that by 2020, 50 per cent of all searches will be voice searches. And Amy Webb wrote that humans talking to machines represents the next major shift in our news information ecosystem, making voice perhaps the next ‘big threat’ for journalism.
How can you reach your audiences where they are?
In a session at the GEN Summit 2018, Kourtney Bitterly, lead at The New York Times Research & Development; Raisa Sufian, Business Director at Vayner Media; Mukul Devichand, Executive Editor Voice at the BBC; and David Tomchak, Digital Director at the Evening Standard, discussed how their respective companies are experimenting with voice AI, whether there is an effective way to drive monetisation via this medium, and why you shouldn’t be afraid to enter the market.
Be a first-mover
When it comes to using voice assistants for news, the focus isn’t on the money just yet. At the BBC, The New York Times, and Evening Standard, the teams are mostly just playing around with the format, because publishers need to understand and be present where their audience is. ‘If not, you risk losing relevance’, Sufian told us. ‘And as a first-mover, you get the opportunity to be the go-to news source for your audience in a whole medium’.
Research, research, research
According to Bitterly, in order to understand how smart speakers are factoring into people’s lives, The New York Times embarked on a month-long intensive research process with audiences in Seattle, Miami, and Detroit. While she said that the team experimenting with Google Home and Alexa at The New York Times have often found themselves being frustrated with the limitations of the platforms, users are generally more optimistic. ‘I think the most surprising thing was how forgiving people are, because they have optimism that it’s going to get better’, said Bitterly.
According to the research, one of the perks of personal assistants is that people feel more empowered to direct their own news experience and decide when they want to let information into their lives.
Out of the twenty people the paper interviewed, every single person mentioned that they had disabled their breaking news notifications.
‘We need to actually start thinking about how perceptions around news are shifting and what that means for our experiences across a number of touchpoints’, said Bitterly.
Experiment, experiment, experiment
The Research & Development section at the BBC also did some research into how users respond to interactions with Alexa skills. (A skill is a capability of Alexa. Alexa provides a set of built-in skills and developers can use the Alexa Skills Kit to give Alexa new skills. An action is the equivalent for Google devices.) Their experiment was an interactional sci-fi drama called The Inspection Chamber, which encouraged listeners to take part in the story by responding to comments and questions from characters. Some of their findings included that users like talking to machines and that they’d like to be more personally implicated in the story. 63 percent of listeners enjoyed the novelty of the format, although 67 percent didn’t like the story. You can find the rest of their results here.
‘What we’re doing right now is just the beginning’, said Devichand. ‘I think at the BBC, we have been technology leaders in the past with video on demand and we are getting into podcasting. What we perhaps have taken a while to do is master the creative storytelling potential of some of these new platforms’. The BBC will continue to experiment with different models and experiences and start creating content that is native to the voice environment. According to Devichand, one area they’re particularly looking at is content for children by experimenting with different play patterns.
The elephant in the room: monetisation
- Amazon doesn’t allow for Alexa’s voice to read out ad copy. The company has not yet found a way of adding advertising on Alexa without compromising the customer experience.
- Voice-based search is tricky, because unlike a website page that lists many search results, Alexa can only present one result at a time. ‘For instance, if an individual asks, “Alexa, how do I cook steak?” If Alexa only surfaces a paid option, it may come across like a hard sell. If Alexa recommends a mix of paid and unpaid yet relevant options concurrently, it would be too cumbersome for users. Either way, it will impair the Alexa user experience’, Michael Nicholas, co-founder and partner of artificial intelligence agency Born, told Digiday.
How can you make money with voice AI, then?
Distribute existing content
Being a public broadcaster, the BBC’s main concern is not monetisation, but there are lessons to be pulled from what they’re doing. Devichand told the audience that the BBC Alexa skill serves as a way to give carriage to great broadcast and podcast content that they have already created. The skill has shown very strong metrics, according to Devichand. ‘I think we had about a million unique browsers in the first six weeks in the UK, which we’re very happy with’. This was relatively easy, as users were looking for something that they already knew they liked: BBC radio and audio content. ‘Discovery of anything else on this platform is a huge issue’, added Devichand.
The New York Times has also put their short and daily podcast The Daily on Alexa devices and Google Home. ‘It’s probably the most successful product launch of maybe the past few decades’, said Bitterly.
According to Digiday, a number of radio stations simply add the listeners tuning in through their stations’ Alexa skills to their total digital broadcast audience numbers. The ads in these broadcasts are already baked into the stream which will run through Alexa, so Amazon doesn’t block them. (The New York Times doesn’t count the Alexa audience as part of The Daily’s overall reach and therefore doesn’t monetise those listens.)
Another way of getting around Amazon’s ad restrictions would be to take a leaf out of The Washington Post, HuffPost, and Bloomberg’s books and make money by having the host read out the ads themselves.
The Evening Standard has extended its GO London section, a curated daily guide to the latest events in London, to Alexa. The GO London skill for Echo Show (Alexa-enabled device with a screen) includes a 90 second briefing, which mirrors the tone of the Evening Standard. (‘Vibrant, open, cosmopolitan, positive perspective — we support the arts and are pro-business’, Tomchak told us.)
‘[Echo show] is a very strange format but we tried that as a bit of an experiment because it’s even more open in terms of the lack of competition and the potential for growth’, said Tomchak.
The aim of the skill is to drive traffic back to the Evening Standard’s podcasts and site, in order to bring about monetisation.
The paper is planning to create showbiz content and news content for personal assistants, but according to Tomchak, there’s still some audience analysis to do, such as identifying the topics people are interested in and the frequency with which the content needs to be delivered.
‘We use data from our platform as well as social and search insights. For any new product we also look at the resource and distribution requirements’, said Tomchak.
‘The benefit of the fact that advertising doesn’t really exist [on Alexa] is that it’s forcing us to forefront the user experience’, said Bitterly. She explained that a lot of skills and actions on these devices exist in two buckets: they’re either highly functional, such as listening to the morning news, or they’re more novelty, such as playing games. The New York Times has been thinking about whether there is a way to bridge those functional and novelty experiences in order to bring about monetisation.
For example, Sufian suggested that if a company owned a voice skill around cocktail recipes, they could make money by featuring recipes sponsored by brands.
Bitterly also pointed to the example of Chompers, an Alexa skill by Gimlet. It’s a twice-daily show that accompanies kids while they brush their teeth. The first season is sponsored by Oral B and Cresta Kids. ‘With this move, Gimlet has made the choice to dive headfirst into the ethical hairiness of advertising to children’, wrote the Nieman Lab.
LEGO has also released an experience that is closely related to a product. The Lego Duplo Stories skill came out earlier this year and it combines storytelling with brick toys: Parents and children are given audio prompts in order to accompany physical play.
Bitterly told us that at The New York Times, they already have T Brand Studio, their in-house branding content studio, which may be the place where they’ll start building skills and actions for other brands in order to make money from these platforms.
Tomchak hopes that voice will go beyond search and that users will one day be able to interact properly with their devices through interruptions. This would allow for affiliate sales on the GO London skill, where listeners could say ‘pause Alexa, I’d like to buy tickets to that event’.
Go on, give it a go!
Sufian pointed out that a lot of organisations think that voice is yet another channel in which they need to invest a lot of people and money. But fear not, at the Evening Standard, there is not one dedicated team working on voice interfaces. According to Tomchak, The AV team is in charge of coordinating the work, but the GO London team themselves are responsible for the content that goes out through the Alexa skill.
Bitterly added that from a technical standpoint, it is fairly simple to build skills and actions, making it a fairly easy space in which to prototype and experiment. Devichand added that voice is a very nascent space: not one news organisation has a very advanced creative paradigm that is native to these platforms, so it’s still a pretty equal playing field.
‘At the risk of sounding sycophantic: we [the Evening Standard] are tiny but I am sitting on the stage with the BBC and The New York Times, which says a lot about what this space is like at the moment’, said Tomchak.
While the barriers to entry are low, Devichand did remind us to proceed with caution. In some ways, voice is like the dawn of social media, ‘It’s exciting and creative, but it’s bound to have some quite disruptive effects — in a good and bad way — on the media marketplace’.
‘Social media had a disruptive effect on our business models, on our relationships with our audiences, and on society and democracy. It’s worth thinking about this technology in an expansive way: on the one hand it’s exciting and the barriers to entry are low, but it’s also worth being deeply aware of the quite rapid pace that it might pick up’.
Read more about how the BBC, Bayerischer Rundfunk, and the Financial Times are experimenting with voice AI here.