In 2015, Prisa Radio, the world's largest Spanish-language radio group, began putting in place a digital transformation plan to identify business and editorial opportunities for the future.
As part of this plan, the organisation started thinking of ways to improve audio discoverability in search and recommendation engines, which are primarily optimised for text.
Prisa Radio produces 39 radio formats across 1250 stations, and has 24 million listeners in 13 countries across Spain and Latin America. The network also uploads live and on-demand audio and video content across its network of websites. Out of its 36 million monthly online listeners, 70 per cent come from mobile, while 50 per cent find the content through third-party platforms.
They came up with Hertz, a feature aimed at on-demand audio content available online on the Prisa Radio-owned Cadena SER and through Cadena SER's mobile apps.
"Audio is a 'black box' for search and recommendation engines," Relloso told delegates at the event, "and only advanced users download, cut and share audio online.
"Audio also never goes viral, although talk radio is one of the main agenda setters," he added.
Hertz represents a new addition to the video player listeners can find on the Cadena SER website to access radio programmes in full, or snippets from them, which are uploaded about five minutes after they air, on average.
Hertz was not developed primarily to change the listening experience for people, so the player retained its look and feel – the aim was to enrich audio files with transcriptions and tags to allow search engines to index them.
When a listener arrives to the player now for on-demand content, a transcription is shown in the video as the audio start rolling. A user can also share the entire clip or a portion of it, using the text to quickly identify the desired quote, for example if they heard it on the radio earlier that day.
Enriching the audio with transcriptions and tags means listeners can find it when they search the web for the name of a programme, or the name of the host of a particular show. Searching for this can send them to the player on the website, but it can also display results from distribution platforms such as iTunes, which is Prisa Radio's main driver of on-demand listeners.
"The number of users that access on-demand content is growing really fast but the reality is that this is happening because we are sharing that content, which in the past we didn't," Relloso told Journalism.co.uk in a follow up interview.
"It's also important to note that it's not only people going to our site directly, they are accessing our content through audio platforms, for example iTunes.
"They like our content, they search for it, and now that we are making available lots of information, they search more and consume more. And it's also because of the other platforms that are starting to work in the podcasting and audio on-demand world, such as Tune In, Spotify, Deezer – they are making this type of content available."
Another step in Prisa Radio's digital transformation was Podium Podcast, a podcast network launched in June 2016. The digital platform was designed to allow the organisation to experiment with new audio series and shows that don't air anywhere else, and are created and distributed through Podium and platforms such as iTunes, iVoox and others.
The podcasts are about entertainment, fiction and 'essentials', a category focusing on historic events, big interviews and theatre shows from the 50s and 60s from Prisa Radio's archive.
"With Podium Podcast we are moving into new content that is not available on the radio, and in the first year, we had more than nine million downloads."
With the funding from Google's Digital News Initiative, Prisa Radio will work to expand Hertz in the next six to nine months to transcribe and tag all the audio content from its archives, using automatic transcripts.
"In the original thinking about Hertz we thought about the archive of Prisa Radio and the traditional programmes of the organisation.
"We are making it accessible at the same time that we are doing the transcription for people to better find that content and share it."