Ofcom UK Podcast Survey 2021

Ofcom UK Podcast Survey 2021

Thanks to someone on the UK Audionetwork email group for alerting me to this data that Ofcom has recently published in tabular form on its website.

Ofcom has just published the data tables for a piece of consumer research on consumer attitudes and awareness of podcasting. I suspect that Ofcom will be publishing their own report summarising this data, but in the meantime, because the data is there, I thought it’d be worth getting into.

The research was conducted for Ofcom by Yonder, and the technical details of how they carried out the research, and the weighting that was applied to mirror the UK population, and not just the online population, are given in their technical report. The full data tables are here (Note that to fully explore the Excel formatted tables, you will want to “unfreeze” the page). In essence, they interviewed 1,865 people, although some questions were only asked to subsets of those respondents.

I should also note that the charts below are embedded via Google Sheets. This means that hovering a mouse over them should bring up the values if you find some of this hard to read. The downside is that in portrait on mobile, some of the charts will get cut off. I can only really suggest that you ideally use a laptop or tablet to read the detail.

With that, let’s dive into the data.

The data begins with asking about how often respondents listen to different kinds of audio.

This is really a reminder that radio remains the most popular audio source. However, podcasts are at 50% for “Ever” listen, and 25% for weekly listening (“Regular” listening on the chart).

Looking at those “Regular” listeners in more detail, here is the demographic profile of UK podcast listeners.

So regular podcast listening skews male, and it skews young (16-44), although with a non-trivial number of 45-64s. These regular listeners also skew very upmarket, making it a valuable advertising market, but also raising questions about the kinds of podcasts that are being made to reach people in other socio-economic groups.

Of note is the fact that of those who say they “Never” listen to podcasts, 20% of them have listened to one in the past. And looking at the demographics, the biggest group of trialists who’ve then abandoned the medium are 16-24s, with 38% of them having at some point previously listened but not gone back.

Exploring some more those who no longer listen we get the following results.

Not finding anything to interest them, or having enough time are the big reasons here. The Covid lockdown has changed things for some, with 12% of those who’ve stopped saying it’s because they’ve not been commuting, and 4% saying because gyms have been closed.

Digging further into the reasons for people never having listened to a podcast at all:

Lack of interest in the medium is the big one here. But 25% of this group don’t know how to access them, and 23% just don’t know what they are. I’d suggest that some of those who are “not interested” may simply not know what they are either. Respondents don’t like to admit to not knowing things.

The occasional listeners (less frequent than weekly) were asked why they don’t listen more often.

Lack of time and lack of interest are the key reasons here.

The questionnaire then asked a fairly fundamental question about what a podcast in the eyes (or ears!) of listeners.

I think respondents could only choose one of these options, and arguably all of them are “right answers” to one extent or another.

Most podcast listeners are relatively new, with 55% of them only having started using the medium in the last two years. A reminder, perhaps, for those of us who have been listening to podcasts for a lot longer.

Generally speaking that has to be good news, with people listening to more than they did. But a small reminder that our lives do change over time. 26% of people listen to fewer podcasts. So why are those people listening to fewer?

So “time” is the biggest issue (How do I listen to more podcasts when there are all those Netflix shows?), but Covid has an impact too. However, I would say the biggest issue is “discovery.” 24% say that they find it hard to find something, and 21% say they don’t find other podcasts interesting. (Respondents could choose multiple options in this question).

Getting into “discovery” some more…

Again, respondents could choose multiple categories here.

As expected, direct recommendation seems really important here with social media at 31% (which obviously can include messaging from podcast creators as well as friends and people you follow), followed closely by word of mouth from friends and family at 26%.

Radio has a big impact – obviously an audio platform is well placed to promote another audio platform. And the audio services themselves are important, as any podcaster who’s managed to get their podcast featured on an app’s “front page” will know.

The next chart has a lot of information in it, since it compares podcast listening with other forms of audio (and music videos) during different activities.

So music and speech radio dominate in the car, and audiobooks are very popular just before bed.

Podcasts do still do decent numbers in car and on public transport. But walking is the big win here, with mp3 listening only just edging out podcasts. (Incidentally, although mp3s and streaming music are counted separately, I suspect respondents may have concatenated them a little). Podcasts are also very popular during housework and just at home relaxing. They’re second only to audiobooks just before bedtime.

The full data tables break out all these results by various demographics, but I would note a couple of podcast-specific things here. Women are much likelier to listen to podcasts during housework than men (47% v 37%), which may say more about our society than anything. They’re also substantially more likely to listen at bedtime (37% v 27% for men).

Ofcom also asked about why people listen.

Again, there is a lot of information here. Looking specifically at podcasts, 24% say it’s for background listening (No – I don’t understand that, as I really can’t listen half-attentively to most podcasts!). 19% say podcasts are for company, and 37% find them relaxing. 21% get practical advice, but the biggest category all round is entertainment. More people get entertainment – 61% – than any other audio category (“The most entertaining audio”: Ofcom research).

21% say podcasts give them something to talk about with friends and family, and 23% use them for news (although speech radio dominates here).

That’s an average of 6.2 episodes a week (with a standard deviation of 0.23). Men and women are pretty similar, with men averaging 6.3, versus women averaging 5.9. In terms of demographics of those averages:

You can see that the biggest podcast consumers are actually 45-54s. But I find it really interesting (and encouraging) that across socio-economic class there’s a lot of consistency.

Next the survey asked about the number of podcast series people subscribed to (Obviously this was all done before Apple decided that “subscribe” meant something else, and changed its terminology to “follow”. But that’s a whole other issue!)

The mean number of podcast series subscribed to is 9.01, but you can see that a hardcore 8% listen to more than 20 series. I suspect that many readers of this piece will, like me, be among that number.

The really interesting thing here, and perhaps most concerning, is that a full 10% of podcast listeners don’t subscribe to any podcast series at all. Either they don’t understand that functionality, or perhaps that’s just not how they hear podcasts.

Oddly, Ofcom also asked respondents how many of these series are “active” and currently releasing podcasts. I’d say that’s a very hard question for the average consumer to know. Sometimes a podcast series has a definitive end, and you might well delete that series entirely from your device. Or it might bubble away in the background, being used from time to time to seed promos to other podcasts from that provider. On the other hand, a series might go away for a period of time before a new season is commissioned. The average consumer is unlikely to know that. Goodness – I work for an organisation that has such series and even the people who work on some of these podcasts don’t know if they’re getting a new season or not!

Anyway, for completeness, here are the numbers.

Respondents were asked what proportion of individual episodes they listened to.

Kudos to the 21% of people who listen to everything. I fear that I’m in the 29% who said “some of them”. I worry about the 2% of people who say “none of them” and perhaps don’t know how to unsubscribe?

The next question got into different categories of podcasts, and seeing how frequently people listened to different types of podcast. I’ll summarise this with three charts.

The first chart shows people who “Ever” listen too podcasts in each category.

These numbers seem to in general terms agree with what other research will have told us. Perhaps the only really new information is the 42% of regular podcast listeners who have listened to Covid-19 podcasts.

When you start looking at people who listen at least weekly, then the big categories begin to stand out more.

Finally, here is the breakdown of listening for each category by each frequency.

News/Current Affairs and Entertainment are the two standout categories here for frequency, the former perhaps unsurprisingly.

The full data tables give demographic data for each type of podcast, and I’m not going to chart all of that here. But a couple of nuggets I spotted:

  • Fiction podcasts skew slightly female, with 57% of men saying they ever listen, but 65% of women saying the same.
  • If you want to reach 16-24s, then the obvious categories are the biggest: Entertainment (67%), Comedy (72%) and Discussion/Talk (74%) podcasts. But don’t overlook Science and Technology, with 65% of 16-24s listening to these.
  • True crime does indeed skew female (64%), although plenty of men (54%) do listen too.
  • The reverse is not so true for football podcasts which do indeed skew incredibly male (61% v 30% female).

Overall, comedy wins the popularity contest.

Although breaking this down by gender reveals some significant subject differences.

And here’s the same data by age group, which does get harder to interpret. But there are interesting things to spot like elder listeners loving drama podcasts, while younger listeners love educational podcasts.

Ofcom asked about what respondents had done as a result of listening to a podcast.

And then they got into the weeds in asking about what apps people use to listen to podcasts.

This is where the sample size for this research doesn’t really help, so use this data with caution. Nonetheless, it’s very interesting in that it shows Spotify at the top, with BBC Sounds very close behind. Apple is solidly in fourth place which I find interesting.

It’s also important to point out that this data is for all regular podcast listeners, but doesn’t reflect the number of podcasts listened to by those people. So if I listen to one podcast a week on Spotify, I count the same as you listening to 25 podcasts a week in Apple Podcasts.

Note that Ofcom’s research got right into the weeds of different flavours of free/subscription/trial of Spotify and so on. See the full data tables if you want to get into that.

They did also ask some specific questions about the websites of radio stations, newspapers, and those of podcasts themselves, but you can look up that information yourself. I didn’t find it enormously useful.

Perhaps more interesting is examining how many different platforms people use to listen to podcasts.

Most use either 1, or perhaps 2 platforms. But a handful use an awful lot. With moves towards platform exclusivity for some, will this change over time, or will people not feel the need to use additional platforms to find podcasts?

The overall mean number of platforms people said they use is 2.9.

There was also a question about devices used:

It’s definitely worth noting that 19% of podcast listeners do not use a smartphone. (Ofcom actually has details of device listening by application if you really want to dig into this data.)

They also asked about features of podcast apps and how important they are. (Note that I’ve simplified the 5-point scale down to a 3-point one for clarity.)

The most important things are the quality of the podcasts offered and the ease of use. Consumers like free apps too, while the range and navigation are useful.

At the other extreme, exclusive content scores relatively lowly, as does the ability to share clips socially. Push notifications aren’t enormously important either (Too many notifications?).

I’m not digging into the detail myself, but each of the main podcast apps’ users were asked about that app’s performance for each of these criteria. Again, beyond the bigger apps, this will be less relevant because the sample sizes are essentially too small.

Smart speakers are prevalent these days, and Ofcom asked those who don’t use them why that was.

Unsurprisingly, not owning a smart speaker comes out top. But the fact that so many people listen via headphones is a close second.

Because Ofcom is the UK’s broadcast regulator, they were particularly interested in podcasts that have some kind of relationship with broadcasters – be they radio or TV stations.

Ofcom was interested in why people listen to podcasts on YouTube in particular.

They are also interested in how podcast listening impacts on some of consumers’ other media consumption.

Most people claim it hasn’t affected what they do, although 31% do say that they listen to less of their own music, as opposed to only 19% saying they listen to streaming music services less. Indeed 30% say they listen to more streaming music as a result. And 36% say they browse the internet more as a result of podcasts. Perhaps that’s looking things up? Or finding something else to do while you listen?

Ofcom asked respondents to agree or disagree with various statements. I’ve simplified the 5-point scale to 3-points again, for clarity.

People claim to find podcasts easily and that there is much more choice compared with radio. They love convenience of them too.

They’re not too worried about the content of podcasts being offensive or upsetting. And most people don’t want to start their own (despite what lockdown seems to have brought us!).

50% say they’re not only interested in podcasters from celebrities/influencers. And 59% appreciate content warnings about offensive language.

35% say they’d pay for their favourite podcasts, while 39% say they wouldn’t.

37% say there are too many ads, while 32% don’t believe that to be the case.

And that’s the lot! As I say, there is much more detail in the tables, including some geographic and racial breakdowns along with the demographic breaks.

One final note. It’s very tempting to compare this data with other countries’ data – Edison Research for example. Or even compare the numbers Ofcom have published with those that RAJAR have previously published. But that would be a bad idea, because different research companies use different methodologies and different definitions. Even something as basic as the overall base for the country in question can vary. So comparison numbers can be pretty meaningless. It’s a bit like comparing Netflix’s numbers for a show (“So many million watched at least two minutes of one episode of Bridgerton”) with BARB’s numbers for a UK TV show (“So many million watched last night’s Line of Duty on average across its 60 minute duration”). They really can be like comparing apples and oranges.

All errors are mine. Please let me know if you spot anything!






6 responses to “Ofcom UK Podcast Survey 2021”

  1. Mike Hanson avatar
    Mike Hanson

    Really interesting and thanks for going through all the data so I don’t have to. One question that really perplexed me was ‘Where do you tend to hear about new podcasts?’ No one said ‘from other podcasts’. Very weird of Ofcom NOT to have that option in the question. I’d say the majority of new podcasts I’ve discovered has been because it was advertised on one I listen to regularly. And it flies in the face of the main marketing strategy of ‘If you want to publicise your podcast the best way is to be a guest on another podcast.’ And given how may podcasts are advertised on all podcasts these days it seems, everyone in podcasting thinks so.

  2. adambowie avatar

    Thanks for the kind words.

    Yes – I think it was an oversight of Ofcom (or their research partner) not to include “other podcasts” as an option for that question. It’s certainly the case that as we see the growth of more podcast networks, most take up the opportunity to promote sister podcasts from the same group in their podcasts. It’s a tried and tested formula, and it usually free!

    And then there are things like “Acast Recommends…” which are essentially cross-promotion of otherwise un-related podcasts across podcasts that share the same advertising partner. You even see new podcasts paying to be advertised on appropriate podcasts.

    I think this is the most comprehensive podcast research I’ve seen Ofcom do, so I’m sure they’ll take on learnings for next time they do it.

  3. James Cridland avatar
    James Cridland

    Fascinating stuff, which will be reported on today in Podnews’s newsletter.

    It would be very helpful to the industry, I think, for Edison Research’s Infinite Dial to be undertaken in the UK. We currently have, as you note, no comparable data with other countries; and that’s a shame.

  4. adambowie avatar

    Thanks for the write-up.

    Yes, it’d be lovely to have consistent global research that would allow for comparisons. Ironically, it might be because Ofcom does this work and publishes the results free of charge, that it’s harder for a company like Edison to come in and replicate their other work.

    Ofcom has its own reasons for doing this research, in particular those questions that relate to broadcast media and the confluence with podcasting, and they are duty bound to make their work available. The questions that they asked here were actually repeats of some 2019 research, but they used a different research company this time around, and the establishment survey had slightly different methodology.

  5. Hayley Donaghy avatar
    Hayley Donaghy

    Thank you so much for doing this Adam, this information is so valuable and SO much easier to digest the way you present it.
    Did they report ‘How often do you listen to the following’ by age? Would be interesting to see how that table looks across the age groups.
    Thank you again, really appreciate it.

  6. adambowie avatar

    Hi Helen, I’m glad you find it useful.

    All the questions are broken down by age, gender, socio-economic class and more. You just need to dive into the Excel files that are on Ofcom’s site.

    It’s a very long table, but you can see every answer broken out that way. For example, in row 23,088 (!) you can see the age breakdowns of different frequencies of listening to podcasts that have videos. It’s just a question of ploughing through the data. Good luck!