The radio industry, in the next few years, will face even greater challenges from podcasts. In part, it’ll be because podcasts will be easier to consume. For example, imagine being able to jump in your car and by simply saying, “Play Balls Radio” you will find 15 minutes of side splitting satirical commentary to send you on your way. Or by asking for “The Morning Call” you’ll instantly be updated on the state of the financial markets. The second factor is, though, that podcasts will become more accountable than radio. We’ve seen this phenomenon across the digital space; advertisers don’t know how many people turn to their ad in a magazine, but they know, in theory, how many times their ad is served online. Podcasts will increasingly have the same influence because listener metrics will be so much better than radio audience figures.
Radio figures are a bluff
How do we know how many people listen to the radio? We ask them. A bunch of people roam the streets leaving pamphlets in people’s homes, asking them to fill them in when they were listening and to which station. It’s a process beset by weakness. For example, you’ll remember Radio 2, more readily than a small station whose name you haven’t grown up with. So big station’s always win. But, similarly, people might forget to fill it in, or assume they listened longer than they really did. And it relies on a significant sample size, given the complexity of the data being gathered. It’s not a binary question, like asking whether you support Brexit or not.
Podcast measurements are improving
Podcasts, on the other hand, measure all listens. Okay, that’s not strictly true. A radio friend once denounced podcast stats because they measured downloads. His point was, you didn’t know how many downloaded podcasts end up never being played. That might have been the case a few years back, when we dare not download on the go for fear of eating up our meagre mobile data allowance. Not such an issue today, with most podcasts downloaded as required, or simply played on a website or app – such as BBC’s iPlayer. So, I’ve been able to see how, over the course of a year and a half, the audience for one of my podcasts has shown a healthy growth, as shown in the chart below (the drops, by the way, are when we take a lengthy Christmas break).
Topline statistics like this show something is going right (or wrong, if the figures are going down). But it's still fairly superficial.
How Apple is helping with engagement
Some podcast hosting companies are now offering YouTube type analytics tools, so you can see how many people are listening and for how long. Apple is amongst them, with iTunes Podcast Analytics, a free capability for anyone featured in the iTunes podcast catalogue. This is a big leap forward for podcasters for three good reasons:
- It helps improve your content. If people are drifting off half way through then you know the podcast is too long or the content is not compelling enough to last the course – actually, they are the same thing; your podcast can be longer if the content is strong.
- It highlights potential. If people are engaged you know you have a hit on your hands. Unless you have exhausted a particular niche, there’s the opportunity to advertise and extend your audience.
- It’s ammunition to use with advertisers. Advertisers might be reluctant to have their product plugged at the end of a podcast because they can be less certain the audience is still there. Hence, podcasts tend to promote sponsors at the beginning, when you really should be focusing on exciting the audience with what’s coming up.
The biggest benefit, though, relates to the first point. I have seen software for radio stations that track the sentiment of a focus group, listening live to a show. The program manager can use it to coach his talent on how their content is being received. Fabulous. Except its just a focus group and its only a select piece of content. Tools like Apple’s podcast analytics allow you to measure this for all your iTunes listeners (okay, those on iOS5 or above), for all your content.
I was pleased to see how well received one of my podcast series was. Less than a quarter of the audience listen on iTunes, but it’s a bit enough sample to gauge how well the content is being received – and you can play the audio in the tool to see the points where the audience starts to fall. In this riveting episode (shown below) at the onset of this week’s financial “correction” almost all the audience stayed with us to the end. In fact the average time spent per device was 13 minutes, for a podcast that was only 12 minutes long – that doesn’t often happen; it means people must have been replaying some content. Hey, there's another benefit of podcasts!
This sort of analysis certainly helps you to fine tune your game. I’ve instinctively been one of those rude interviewers who interrupts when the answer gets boring, but the Apple tool has vindicated me – there’s a clear sign that the audience falls a little with long answers – so interrupting keeps the audience with you. Podcasters can use this tool to see if their intros are too long (they often are), if the whole thing goes on too much (they often do), or whether people are missing out when there’s too much friendly natter without much substance (my podcast pet hate).
The upshot is, podcasters now have this sort of capability to perfect their art. Over time we can expect tools like this to be overlayed with demographic data, just as it is with online display advertising. Meanwhile, radio stars are left with the entrenched attitudes of their content directors who may or may not be in tune with what the audience really wants. So, in the brave new age, where podcasters and broadcasters will be fighting equally for their audience share on voice activated in-car devices, who will have the best idea of what the audience really wants?