The Internet Radio "Super Demographic"

Our portfolio company TargetSpot recently conducted two large research studies on the Internet Radio listener. For those that don't know, TargetSpot is the leading third party advertising network for streaming audio.

I've been an Internet Radio/Streaming Audio listener for over a decade and have always thought that Internet Radio listeners represented an ideal demographic for marketers to reach and influence. The TargetSpot research confirms that and then some. Internet Radio listeners are a "super demographic" and marketers should pay attention to this research and start creating programs to reach them.

Here are some of the findings:

– 39% of the US population listens to Internet Radio regularly

– Internet Radio listeners are affluent and influential

– 80% of Internet Radio listeners spend between 1 & 3 hours per day listening

– 45% of Internet Radio listeners listen on a phone and 14% listen on a tablet

– 73% change stations throughout the day. Internet Radio listeners are not exclusive to any one service

– 56% listen to Internet Radio while shopping

– running an Internet Radio campaign in parallel with a broadcast radio campaign increases response 3.5x

– running an Internet Radio campaign in parallel with an internet banner campaign increases response 2x

The way I see it (confirmed by this research) Internet Radio listeners are the leading edge/cutting edge demographic that try out new things first, recommend them to their friends, and provide the word of mouth mojo marketers are looking for. And they are available to be reached on average a couple hours every day. That's a super demographic to market to.


Comments (Archived):

  1. William Mougayar

    You can count me inside most of those stats.So, is advertising the primary revenue model for the burgeoning Internet radio stations? And how are they doing? I almost feel we’re getting a lot of free usage. I’ve only paid for one station, and others I listen to have very light and sparse advertising.

    1. Philip J. Cortes

      Depends on the medium you frequent – it’s my understanding that digital version of terrestrial radio stations play more advertising than say, Pandora.

  2. Harry DeMott

    And now what we need is to see better 3rd party measurement so that the medium starts to fit into the proper boxes at media buyers. CPM’s are still way too low given the data you cite – plus the wealth of demographic and location based data you can target with. I figure given all the data – you should command 2X+ the CPM’s of terrestrial radio.

    1. Tereza

      Amen, Harry.As a media person this is useless to me. No gender and age — huh? If we don’t have them, then I’m not sure who cares.Also radio is such a totally fragmented media buy. I’ve talked to dozens of radio buyers in the past, it’s such a tough and thankless job. I can’t see them making a decision on it but they have the purse-strings.The buying agencies are extremely rigid, almost ‘manufacturing like’ in how they buy under their given goals. They are young people who are not given much flex to think outside the box.Also a massive, massive pain point for them is the paper and invoicing, which is insane. It is as much a factor in who they decide to buy, because the dollars in a single buy are small.

      1. fredwilson

        We’ve got gender and age. I just didn’t think it was a particularlyinteresting data point given the numbers I saw

    2. fredwilson

      I agree. But that will take some time to happen

  3. William Mougayar

    Coincidentally, Accenture released a new study today on Internet video, and they found that “Consumption of video over the Internet is no longer a millennial-generation phenomenon; it’s an activity that crosses all age groups”. So, age is not even a consideration. I wonder if the same applies to Internet radio.Consumers of all Ages are Going ‘Over-the-Top’:

    1. Scott Nixon

      My 65 year old father is more into Internet Radio than I am and was also an early adapter to Satellite Radio. His has never been able to figure out the artists he likes or remember the names. With Internet Radio he is able to curate the channels to his taste.

  4. Sebastian Wain

    The 45% that listen on the phone and 14% on the tablet are using WiFi and/or a mobile network? do you have numbers specifically for mobile networks? since I think this is a very different usage (WiFi is more “in situ”).

    1. fredwilson

      I do not believe they drilled down that far into mobile behavior

  5. Andrew

    So wait, according to this survey nearly one third of Americans listen to 1–3 hours of internet radio every day? I’m sorry, but that’s just silly talk.

    1. CliffElam

      That struck me too. I figured I’d misread it.My son, however, may make up for it with Pandora – he is a podcast junkie._XC

    2. Niyi

      1-3 hours a day sounds plausible but I’m not sure about a full third of Americans listening for that long.Perhaps their definition includes cloud-based “radio-like” listening services such as Pandora, Spotify, and not just Sirius XM.If you view it that way, then the numbers make sense.Either way, inserting “targeted” audio ads into audio streams seems like a winner.

    3. Andrea Carini

      I think the survey results have been reported incorrectly by Fred. 39% of households reached by broadband listen to online radios. I believe broadband reaches 65% of the US population thus is more like ~25% of the overall pop

      1. Austin Clements

        Actually, a step further, the report just says 39% of broadband households listen to “Digital Audio” which includes the subset that is “internet radio.” Digital audio by itself also includes mp3s and digital broadcast radio.Now the listening time is another story… That just seems a bit off in my opinion.

        1. Satish Mummareddy

          I think a lot of people play music in the background while working, doing chores etc and that makes up the listening time.

      2. fredwilson

        I’m sorry if I messed that up. It was not intentional

    4. ShanaC

      It wouldn’t surprise me on the other hand if that number became true – only because think of all the places where you want to be listening to music

  6. Satish Mummareddy

    Most of the research makes sense. But there is a reason the CPMs might be low: during a majority of this time people are usually working (or doing something else) while playing music in the background and tune out the ads.

    1. fredwilson

      I ‘tune out the ads’ when I’m in the car too but then I realize I know theads by heart a week later

      1. Satish Mummareddy

        I can believe that. You need to subconsciously hear the ad a number of times for it to stick.My point is that the cost per recall is the same but the cost per ad play is lower as you need to play the ad a number of times to achieve recall.

      2. Tereza

        That definitely sounds like a heavy frequency buy.You know what they should do, have parents pay for streamed SAT-word “ads” heavily sprinkled into their own kids’ audio streams, at whatever level their kids is at.Studying by osmosis.Kinda like baking zucchini into their brownies. Or — Schoolhouse Rock by Satellite!

      3. mdudas

        Data shows that click-through rates on online audio ads are much higher than click-through rates on standard display ads. For this to happen, the advertiser must ensure that each audio ad has a call to action encouraging the listener to click on the banner, and they do so at a pretty high rate. So, based on the click-through rates, people actually are listening to the ads and call to action on the desktop and mobile internet radio services and then acting upon them.I’m sure that TargetSpot is seeing some of these same engagement metrics on the ads their clients are running that have a call-to-action that is appropriate for the interactive environment that desktop and mobile internet radio provides.Bingo on your point about in-car ads. Particularly when the ads are on services with a relatively low ad load, the recall will be there.

        1. Satish Mummareddy

          Thats interesting.Is there data on time of the day relationships with click-through rates?Age with click-through rates?Age & time of day with click-through rates.In some sense, a relationship between what users are doing in parallel while listening and click-through rates. 🙂 That would be very useful to planning.

  7. im2b_dl

    not to mention content creators should be looking at how many people are looking at their phones/tablets/computers…while listening to their televisions.

  8. Trey

    Not advertisement related but I love KCRW.

  9. Kevin Drost

    Great numbers. It seems most of the streaming consumption is of NPR, non-profit related broadcasts, which is great, but not really a great business. The biggest issue I see facing ad-supported radio streaming is the economics of of per-track statutory licensing. There are no economies of scale. You have to reach a tremendous audience, even if they’re a “super” demographic, which is difficult seeing that 73% change stations throughout the day.

  10. Jim from

    Maybe it’s because I’m young, but everyone I know mutes the sound or takes out their headphones to talk during an advertisement.

  11. kirklove

    Those numbers seem really, really high. Almost too good to be true.

    1. Gkamstra

      Agreed – not as much the 1-3 hrs a day, but 39% of the population? No way… I just don’t believe it.

      1. mdudas

        If this report includes Pandora and YouTube, then 39% of the population could be accurate.

        1. kirklove

          Yes, that’s what I thinking. I need to read the report. But, it’s so much more fun reacting without knowing all the facts. 😉

  12. RV

    I am not a heavy user of internet radio (prefer downloads + podcasts) but I have definitely noticed the people around me using these services heavily. Getting internet radio available in the car during morning/afternoon commutes would be a huge catalyst. TargetSpot seems positioned perfectly. Thanks for sharing the stats.

  13. Tereza

    Interesting at a high level, but since they use the word “demographic” how come they don’t cut it by gender? And also age?Or is it assuming that it just is a heavy male skew so it’s not worth the breakdown.I opened it up yearning to confirm or refute my hypothesis that it’s a heavy male-skew, and hoping to be proven wrong. They skipped over that basic question and I’m dangling here. Piqued my interest but left me unsatisfied.I only did a quick scan so if I missed the breakdown please someone call me on it.

    1. Rayhan Rafiq Omar

      This post, more than others I have read, does seem like a pitch for internet radio advertising.

      1. Dave W Baldwin

        a matter of being transparent regarding useful metrics… data.

      2. Tereza

        I actually don’t mind a pitch. But it doesn’t give the buyer something to say ‘yes’ to. When a buyer is buying on a ‘demographic’ it’s something very different from what they’re calling a demographic here.Lost opportunity. Too bad.

    2. fredwilson

      I have that data. I will double check but I recall that its awfully close to50/50

      1. Dave W Baldwin

        Wouldn’t be surprised over it being VERY close to 50/50. To me, the more interesting demo would be age with a measure from however far back, current and see the next measure. Bet the 45-60 has a noticible gain.

  14. Mike

    This is important data especially in a world where major advertisers are reclassifying their “radio” budgets to “audio” enabling them to be more flexible. Internet radio only represents 4-5% of total ad dollars in $17b us market, but that is up dramatically over last 18 months. Terrestrial radio is still a massive medium but shift is certainly happening, and creating exciting opportunity to rethink audio formatted ads altogether.

    1. mdudas

      According to RAB, digital revenue was 3.5% in FY 10 or $617 mil:…. The majority of that revenue is not audio ad inventory but rather display and other digital ad formats. The majority of the audio ad revenue is local ad dollars sold directly by the radio stations themselves, just as most of the $17 bil is local ad dollars.I doubt Pandora, VEVO and YouTube Music are included in that figure, though, so the overall digital music ad revenue pie is large and growing.

    2. ShanaC

      Why is that surprising – it is long overdue, you can listen to pandora more easily at work than a radio…

  15. paramendra

    What really gets me about this is that this boosts the freemium model. Books, movies could learn. Music itself.

  16. Rayhan Rafiq Omar

    Could you imagine on-demand audio books supported by adverts, served in the internet radio model?

    1. fredwilson


  17. Wells Baum

    So how would you classify Souncloud, Internet radio or just a music tastemaking website?BTW Fred, just a reminder that I donated to on your behalf.You should have received the digital album of 41 + tracks.I know you’re anti-MP3 so if it does you better just stream it from the Soundcloud API on

    1. fredwilson

      I think of souncloud as much bigger than music. It is the place to host andshare the audio you create and/or capture

    2. fredwilson

      Thank you for that donationI will listen tomorrow

      1. Wells Baum

        Great! Let me know what you think.My recommendations for you:Purity Ring – UngirthedYoav – Easy Chair Les Blanks – Straw ManBut then again your tastes are hard to predict!

  18. Abraham J Safdie

    I’d like to give a shout to WFMU, a listener-supported, free-from station that’s really embraced new technology. Though it’s only across the Hudson, FMU’s signal is more spotty in NYC than AT&T’s. But its online presence has been a saving grace: through its clickable playlist archives and live stream [which for some reason always managed to evade my employer’s attempts at blocking], I not only survived eight years at a large law firm, but emerged knowing more about great new, obscure music than ever before.

    1. fredwilson

      That’s one of the reasons internet radio is so great. The signal doesn’tdegrade

  19. Anne Johnson (still in beta) has an interesting approach to this market – “We pay the artists every time you stream their music or video. Directly to them, it’s that simple. After all, if they stop, the music stops.” Paying them from advertising or subscription revenue. Aimed at listeners looking to find new music – new to them, or newly issued. Easy sharing of tracks or playlists. I’m looking into their technology to see how well it scales and to what else it can be applied.

    1. Dave W Baldwin

      Good job. I was hoping someone was going this route, for it will be the big thing not far down the road.

    2. Guest

      Thanks for posting this. Had not heard of it; will forward link on to some friends & family who produce/compose/write music.


      Thanks for the link, Anne. I’ll try it out with my own music.. Also, thanks for posting this post Fred. I had no idea about this huge market that’s just sitting there waiting to be tapped!

  20. Excelforpros

    This article feels just a little bit too biased, I get that you are an investor and hyping them (thanks for disclosure) but the stats don’t seem realistic to me…

  21. Brandon Casci

    I’ve been listening to on-line radio since 1998 or so. I never really followed over the air broadcasters transition to online broadcasting. I went straight for the micro-broadcasters. I found the variety and quality so compelling, I eventually started to building my own radio stations and attracting a nice crowd of followers. It was fun. Eventually I made a platform for this called Loudcaster (, it’s boot-strapped and doing pretty well.But anyway….companies like TargetSpot have been kindly pointing out these demographics for years, but advertisers are still very slow to put serious money into the medium. I recall reading that 2011 might be around $900M in advertising dollars spend on online radio campaigns, with the bulk of that going to big services like Pandora. They may get 10-12% of that $900M. $900M is a small fraction of total online advertising. The big media buys are still over the air TV and Radio.The ad and marketing experts I speak to tell me there are still some challenges still. As more services come online, the ad inventory growing faster than the the ad spending. Also a great deal of media buying offline is done by geographical market. The audience for online radio is global so it doesn’t exactly fit into that paradigm. Pandora has grown big enough to work with that paradigm. They have large audiences in areas like NYC.I think it will all work out though. It’s just a matter of time. Advertisers will come up with new ways of doing things as more traditional media companies (NBC, CBS, Clear Channel, Viaccom…) grow their online audiences, and companies like TargetSpot will be there to help out.

    1. fredwilson

      I totally agree

  22. Rafael Urbina

    At we have been streaming Internet radio for 10 years now to the Hispanic demographic and have found similar trends to be true.

  23. dow hardy

    This will probably get lost in the “lies, damned lies, and statistics” rants, and it is not meant to pile on, but I can’t reconcile the above positives with the often stated reasoning why Fred/USV would not invest in Pandora.

    1. fredwilson

      I don’t like it and don’t use it

  24. Natasha L

    So true! Whenever I’m working on the computer (most of my day in the Adobe suite as a creative), I’m switching between KCRW, Pandora, Grooveshark, and Hypemachine.

  25. sbmiller5

    I’m curious how audio ads align with “monetization should increase the user experience”?

    1. fredwilson

      they pay the soundexchange royalties

      1. mdudas

        Agree with Fred here. Audio ads can help a service pay SoundExchange and therefore stay in business, which is frankly rule #1 in providing a good user experience.Services like Grooveshark that aren’t making a strong effort to generate revenue and are providing services of questionable legality run the risk of being put out of business by the labels – that’s not a win for the user. It’s really not a win for anyone, as it conditions users to expect things for free that can’t economically be provided for free (such as on-demand music).The ad-supported model pursued by the radio biggies and pure plays like Pandora actually have business models built for the long run.

        1. sbmiller5

          We are making a strong effort to generate revenue (and are generating $) and we operate completely legal under the DMCA (I’m happy to send you the copyright law) and are working hard to sign sustainable deals with all labels.Audio ads technically garner a premium CPM, but their total actual CPM is much lower, when calculated against total impressions (which we’re able to serve banner and site skins against, but not audio ads). If audio ads can actually generate a premium when calculated against total impressions, we’d be more interested in pursuing deals. But it significantly reduces the users experience and causes serious decline in legal consumer consumption (which often sends the user to piracy).Consumers are already are conditioned to get music for free, significantly before Grooveshark came along. We provide and opportunity to get it for the cost of consuming advertising.The artists and labels that have worked with us have seen a new revenue stream (70% of our users gave up on piracy or pirate significantly less) and found a valuable source on data of fans and low-cost promotion.If we had millions to spend on Lobbyist like the labels are doing, there wouldn’t be these misperceptions regarding our legality.

  26. FinanceTutor

    I tried target spot based on an earlier post here at avc. It did not yeild any conversions! Albiet at a small budget of $200 (their minimum) with a free voice ad. The sales guys were great and responsive but intended only to sell. They inference was that this is budget was too small and I should plan a larger budget like $1000 or more to see any results. I figured with not even one conversion, it did not make sense to blow more money up. Has anyone tried this. Any unbiased feedback would be great.

    1. Dave W Baldwin

      Use your gut. If you want to see what they can do, tell them to go from no response at $200 to $1,000 is quite a jump. Why have a $200 pkg if it isn’t going to do anything? What kind of target time or whatever do you get?

  27. Judah

    I would pretty much agree. Most everyone I know within my age bracket (20’s to 30’s) uses internet radio in one form or another all the time–parties, work, in the car, on the phone, on a tablet, etc. And since most of use it at work, 3hrs actually sounds low to me.The top we use are:Pandora (best for parties and discovering new artists)Youtube (has the most music by far)Grooveshark (great for making custom playlists)Hypem (great for finding new music outside of what you are used to)SoundCloud (awesome for sharing music without a video attached) (this isn’t very common with my friends… but some kids use it)I’ve always got music going, as do most my friends. I know that I have bought stuff that I have heard or seen advertized with these services and in most cases, I upgrade to a premium version of the service.

  28. Scott Nixon

    Check Out TargetSpot CEO Eyal Goldwerger as he’s interviewed on This Week in Startups Episode #125 (….Interestedly, I just listened to it yesterday and absolutely intend to advertise with them. I’ve used Trada with mixed success but would consider using them with a different type of business.

  29. Jan Schultink

    Today I was working on a pitch saying that the days of the IBM 360 are back again.Reading this, I “hear” that the days of the radio are back.My children will laugh at me when I am 80 and tell them about 2011

  30. jeffullrich

    In the interest of sharing…I own, a comedy podcasting network with about half a million downloads a week. While podcasts are not internet radio, they’re close.We recently launched an advertising survey for our listeners and here are some of the numbers so far, from several thousand responses:On average, how many hours per week do you listen to audio podcasts:1-2: 5%3-5: 25%6-10: 36%11+: 34%On what device:Laptop/Desktop: 25%Mobile Phone: 28%Portable mp3: 44%Other: 3%Ages:21-34: 76%Male: 85%Female 15%That’s just a sample of the 38 question survey. We also received several hundred write in answers regarding monetization ideas and thoughts.It’s a fascinating time and I’m very excited to be in the thick of it.

    1. Aisha, CPO

      Interesting stats Jeff. I’m quite taken by the gender divide – you have a lot of male listeners.Re: podcasts are not internet radio – I think there is an interesting debate here, videos / films online are still called ‘videos’ or ‘films’, same goes for articles, books, newspapers, magazines…. audio is a medium which has somehow become fragmented, forgotten, misunderstood, overlooked – I don’t know?! Call them podcasts, radio programmes, talk programmes, audio programmes – we’re talking about the same thing :)Music as part of radio programmes and the future of this I’m not sure how it will shape and what the advertising opportunities are. I personally get my recommendations through people’s tastes I trust – this happens to be people I know (friends & family) and a DJ who has his own podcast show (http://www.gillespetersonwo…. So I don’t tune into broadcast or internet radio. I like it on-demand and/or personalised so I like what you’re doing with earwolf.comI personally listen to a lot of talk content and feel that it’s very fragmented, beyond the popular programmes (RadioLab, This American Life) – it’s tricky to find good stuff.Going back to the topic – yes! – I think there could be advertising opportunities in audio but before that we need to have some better services for making audio / radio relevant today – with the tech, tools and lifestyles we lead.

  31. Keith Hopkin

    @citadelgradI am 34 and don’t know what the artists are playing on and FM radio broadcast because the DJ never tell you. I still haven’t figured out why. They hammer the same 10 songs into your brain until you break down and google “What’s that song with that kid singing I’ma tell you one time”. Then you realize you’ve been singing a Justin Bieber song in your head for the last two weeks. Then you realize you …..can’t…..get…it…out…of…your…head…..AHHHHHHH!

  32. Leigh Acton

    Love this article

  33. Adrian Meli

    Do we know what % of internet radio is static at a high speed internet connection vs. mobile usage. If everyone listens to internet radio long term on a mobile phone then I suppose those people would require more expensive wireless data plans to accomodate the usage and ultimately it won’t be free? It is the same issue Netflix is going to face on a wired basis I think…

  34. Jon Anxin

    I would say that about 80% of the internet radio listeners are strong “anti-authoritarians”, as well.

  35. RichardF

    +1 I listen to Mountain FM, Whistler sometimes, just because I love Whistler so much!

  36. Satish Mummareddy

    Damn you reminded me of the Junkies radio show from Maryland, DC. It was my guilty pleasure on the drive to work:( 🙁

  37. ShanaC

    It is national grilled cheese day too… 🙂

  38. fredwilson

    And the data shows that is a very significant use case

  39. maxniederhofer

    I listen to KQED when I’m in London 🙂

  40. RichardF

    that’s two votes for KQED here Max, so I’m going to have to tune in.I like SF but have only spent a few days at a time there on my way through to Tahoe.

  41. Charlie Crystle

    best station ever. I think–what else?

  42. fredwilson

    Nope. Just a word from your sponsor. Got to pay the bills

  43. Judd Morgenstern

    I don’t claim to know the metrics, but if my application of transitive logic holds…- 39% of the US population listens to Internet Radio regularly- Internet Radio listeners are affluent and influential>> then 39% of the US population is affluent and influential…??Not trying to nitpick, but if this logic doesn’t hold, then I have a harder time believing the logic at the end of the post, used to ‘confirm’ a few hypotheses about the market demographic.I sincerely don’t think this was done with malicious or deceitful intent, it is just an example how numbers alone don’t necessarily paint a complete picture, and can be misleading. My friend taught a statistics course at U of Michigan titled “The Numbers Don’t Lie, But You Can Lie With Numbers”.As a former PE guy turned Designer/Entrepreneur, I am more comfortable making and validating hypotheses based on a mix of quantitative + qualitative data. I would be interested to hear about the qualitative / ethnographic research that your portfolio companies generate.

  44. Charlie Crystle


  45. Charlie Crystle

    I absolutely believe it. Number 1 reason I drain the battery