Could the experience of listening to a podcast be made even more engaging by receiving personalised complementary information? Is there a difference in how likely people are to text with the hosts of a show if they are listening to a radio show or to a podcast?
Last year, the Center for Investigative Reporting (CIR) developed a tool called Amplify, to text people charts, pictures or videos relevant to a particular episode of the Reveal podcast as they are listening to it. In June and September 2017, the CIR tested the SMS experience on two episodes discussing in vitro fertilisation and Berkeley protests respectively, and they have since conducted two more experiments using Amplify.
In October, Reveal partnered with WHYY, a public radio station in Philadelphia, to test Amplify with listeners of its weekly radio show and see if the tool would work for a broadcast audience, who tend to have less control over their listening experience and often listen to the radio while multitasking.
The episode was about Trump's crackdown on immigration, and at various points listeners would be told by the host to text certain keywords to Reveal to receive, for example, a map of places where immigrants have been picked up by officials to potentially be deported, or a video of a detention center northwest of Philadelphia.
They were also encouraged to share their own questions about immigration and more than 400 people did so. The team will be producing a follow up radio show to answer these messages.
Although broadcast audiences are measured differently to podcast ones, Reveal found the conversion rates for radio listeners tuning in to Amplify were similar to that of podcasts, said Hannah Young, Reveal's director of audience.
Reveal's most recent experiment with Amplify took place in February, when the team made a significant tweak in how they were interacting with people through SMS, and also tested the tool with their podcast audience, radio listeners and people engaging with Reveal on its website and through social media.
The episode that aired on 17 February looked at modern day housing discrimination and disparities in lending in the US. It was part of a larger investigation Reveal produced on this topic, analysing over 30 million government mortgage records and finding that people of colour are more likely to be denied a conventional home loan in 61 metro areas in the country.
Having such a rich dataset, almost down to a neighbourhood level, allowed for personalised information about lending disparities to be sent to listeners when they texted their address to Amplify.
Tweet could not be found or could not be loaded.
"We wanted to test this idea of whether delivering that type of personalised information to people would make them more or less likely to text, as we thought they'd be more likely," Young told journalism.co.uk.
"We also wanted to see if we could deliver the whole experience with one keyword triggering the messages."
People who came to Amplify from the Reveal website, social channels or partner websites had to text 'LOAN' to receive the messages, while those finding out about it by listening to the radio show or the podcast used the keyword 'HOME'.
This approach enabled the team to track which distribution channel resulted in a more engaged audience. They found that people coming in through the website or social media had a lower conversion rate to Amplify than audio listeners, amounting to just over 600 users.
"We saw a lot more people on the digital side use an app that we developed, which looked like a map they could click on to get the same information they would've got through text. It seemed like texting is probably a heavier lift for folks if they're reading online. It's much easier to click on a map or stay on the internet instead of transferring to SMS.
"On the broadcast and podcast side, we had about 3 per cent of our podcast listeners use Amplify, and in previous experiments it had been around 1 or 1.5 per cent," Young said. She added that the personalised information and the use of a consistent keyword both could've played a role in the higher rate of engagement, and that having just one keyword also made it easier for people to catch up if they happened to join in later in the episode.
Towards the end of the episode, Reveal asked people if they or anyone they knew had had trouble securing a loan, and prompted them to ask any questions they still had. About 1,500 people wrote back, some 17 per cent of the total number of people who had started the messaging experience.
"When we do that in different channels and in other ways we usually see something under 10 per cent, closer to 5 per cent. So it was good to see that once we hooked people in, they were likely to have questions and keep interacting towards the end of the experience, when we were basically saying 'we've given you all this context, now how does that really affect you?'"
Feedback also showed that listeners wanted to use the service to get information about lending disparities in areas other than their own. In upcoming shows, the aim will be to figure out how SMS can be used to clarify complex topics for listeners, or "help them dig a little bit deeper into something that's harder to do in audio format".
The team is also exploring ways to enable listeners to access all the assets sent for one episode in one place, to receive all of them at once or to receive a particular one at a later time in the episode if they are unable to text the keyword as soon as the host mentions it.
"We've now done four of these tests and ramped up with each one, so we're looking at whether this is a niche product for a specific group of folks in our audience, or if we want to keep increasing the number of people using it," Young said.
Free daily newsletter
If you like our news and feature articles, you can sign up to receive our free daily (Mon-Fri) email newsletter (mobile friendly).
