From Scans To Listens
For years the number that musicians and their business partners, managers and record labels, would focus on was the soundscan results. From the soundscan website:
Sales data from point-of-sale cash registers is collected weekly from
over 14,000 retail, mass merchant and non-traditional (on-line stores,
venues, etc.) outlets.
My friends in the music business would always say things like "we scanned 25,000 units last week". In a world where the sales of CDs (and before that albums) was the key goal, it made sense. Scans were revenues.
But the world has changed and more and more music is available on the Internet for free or via subscription services every day. And just because you scanned a record, doesn't mean you listened to it.
I think it is time to stop focusing on scans and start focusing on listens. But how do you do that?
I bumped into my friend Spencer Hyman yesterday at breakfast. Spencer used to run last.fm. I asked him if you could use last.fm as a panel, like Nielsen or comScore, to measure music listens. He said you could but you'd need to do some statistical weighting by geography and genre.
If you look at the last.fm charts for 2009, you'll see that Lady Gaga was most listened to artist on last.fm with 755,000 different listeners and 18.5mm "album scrobbles". If you go to the Lady Gaga page on last.fm you'll see that her song Poker Face has been listened to 235,000 times in the past six months.
If you think of the people who scrobble their listens to last.fm as a panel, then you can scale up these numbers to get to worldwide listens. Last.fm hasn't done this work but they should. They could be the new soundscan as the key metric moves from scans to listens.
As an example of why this metric will be increasingly valuable, let's go back to that Lupe Fiasco mixed tape that includes the HP advertisement as a track in the mix. That advertisment track is called HP Skit. Last.fm says HP Skit was listened to 3,850 times last week. Let's say 50x is the right multiplier for this artist and song (it's an educated guess). Then HP Skit was listened worldwide about 200,000 times last week. If that ad is worth a $20 cpm, then Lupe could have earned $4000 in ad revenue last week.
All of that is hypothetical to some degree but hopefully instructive. Music is moving from a physical good where scans is what matters to a virtual good where usage and engagement matters. So let's start measuring it correctly. That may help the artists get paid correctly.
Also, the music video of pokerface has been viewed 680,000 times on youtube. And that’s only the official one. The unofficial videos visible in the “related links” column have been viewed >200,000 times each.http://www.youtube.com/watc…
I believe this is the model that is used by the music services like Rhapsody – artists are paid, pro rata, by the nuber of listens. I am a big fan of music subscriptions (I use Rhapsody) and I think their biggest problem is they don’t market their key point of differention – you don’t pay for each “scan’, and the subscription model delivers a lot more value for a music lover like me -http://bit.ly/8h7eR3
i’ve been a rhapsody subscriber since the late 90s. i love it.
i’m still dissing CPC/CPM pricing at this point in time and new models dependent upon it. IMHO nothing new and disruptive will come out of something rooted in CPC/CPM; for starters it is too easily gamed. i think at best we’ll see something that sorta looks new and disruptive, but when push comes to shove, it will make sense for that thing to get acquired and be assimilated into some pre-existing borg and that borg’s value chain. breaking out of CPC/CPM pricing is what will truly change everything, and personally it is the one of the things i am most interested in seeing — who can create the non-CPC/CPM ad network, and what such a network looks like. based on stuff so far i’m inclined to think it’s closely related to social media and social gaming. IMHO the non-CPC/CPM ad network has the potential to be the new economic keystone that drives the web.in sum i do think the analytics data has value, but i don’t know where it fits in to the overall picture in this scenario, and would prefer to see a non-CPC/CPM pricing model emerge to dictate how analytics will fall into the picture.
These new concepts will not be ads networks anymore (a good thing). Anything you push will always use CPM, and ads are by definition push oriented. It should be new ways that adapt to consumers real needs rather than shoving whatever you have to sell now. Music consumption is a great place to start.
The internet and those that benefit from it are hungry for growth. I agree that older ad measures are broken, but they won’t be fixed until WE fix them. That means you have to invent a monetization method connecting social media to businesses that’s effective, and even desired by users.Totally agree we gotta shake this system up. Look at social gaming, they just skipped ads and went straight to cash, although they have “watch this ad for X points” options.
And even to separate “pull” vs. “push” listening.There is a difference between actively asking to listen to a specific artist/song vs. getting it in the suggestions stream of Pandora or Last.fm. In fact, this is a measure which is closer to buying a CD than listening to radio.
Great detail. User action and behavior should be connected to the measure.
I definitely agree with the principle, but a chart based on listens would appear to be a lot more vulnerable to hacks, scams or generally being “gamed” than scans where a financial transaction must take place. Money changing hands is a very powerful factor in determination of value.
good observation about “gaming” the numbers.however, money changing hands is not the only determination of value. i’ve bought quite a few cd’s or individual mp3’s based on 30 second previews and maybe listen to the song two times after that. other songs i’ve purchased are in my “top tracks” on last.fm and will be there for years to come. listening to 4 minutes of music many times says a lot about how someone values a track.
Until there’s a way to turn attention into revenue, listens is an interesting but largely meaningless metric.Attention is not the problem. Music gets lots of attention. Measuring attention is important (back in 2000 I had a company called Songstats that did a lot of what everyone is talking about now). But without knowing the formula to convert listens into dollars we don’t know what we’re measuring.Be careful what you measure, you may just optimize for it.The HP example is an interesting case, but ultimately I don’t think we want product mentions in songs funding the music industry
Optimizing for what you measure is a great supporting reason for Colm’s focus on transaction.Sounds like Last.fm could use your expertise in this space if they do seek to characterize listens.
the future is all about monetizing attention not the transaction. it’s time to start focusing on that.
I agree.As usual I was 10 years too early :-).Sometimes the world isn’t ready for what you want to do. You know how hard I pushed to be able to sell users more chips for cash money in blackjack and poker in Y! Games?
Focusing on engaging the individual rather than the transaction drives monetization I believe for services. You are saying that for music, attention equals engagement. Makes sense.
Interestingly enough, Paul Graham sees iTunes etc as “tolls” – it’s all about the transaction and not the content itself.I think he’s right.Good read either way for anyone who missed it – Post Medium Publishing http://paulgraham.com/publi…
Did miss that one (day job eating up all my cycles), thanks Reece.*edit* Whoops, I caught that one, and yes it’s certainly a good read. Paul does a great job of capturing his ideas and communicating them “optimally”.
I dont understand what you mean. Monetization IS the transaction, no? How does one monetize attention without some form of transaction?
“the future is all about monetizing attention not the transaction.”If the attention doesn’t sooner or later lead to a sale, then I would say attention in and of itself doesn’t have a monetary value.No matter how much attention people are paying to music, if they aren’t (1) buying the music, (2) financially supporting the band that created the music, (3) paying for the service that brings them the music, OR (4) buying from the advertiser or sponsor that underwrites the music, then monetization hasn’t been accomplished.
I agree particularly to the last point. But that’s not a music transaction. I believe music attention leads to powerful monetization non music transactions. Radio is a 20bn business offline
“I believe music attention leads to powerful monetization non music transactions.”In order to measure monetization of non-music transactions, businesses try to monitor how much their advertising and sponsorship dollars lead to sales. So even if they support music that gets lots of listens, that in itself isn’t enough of a measure for them. They are going to look for ways to see if that association with music results in measurable sales.So a better measurement for them isn’t listens, but click-throughs leading to sales, coupon submissions, and other types of direct response. If there is a way to link listening to music directly to listening to sales, that’s the measurement marketers want.A limitation of radio is that advertising on it usually isn’t as directly linked to sales. As media becomes more interactive, companies are gravitating to media that can be directly linked. For example, announcing a sale only on Twitter and then measuring sales is a better measurement for them because they can then seethe result. I suppose they can run an ad only on a YouTube video and then see what happens.At any rate, monitoring listens only tells an advertiser how many people are listening to a song. Better demographics might present a profile of who is listening to the song and what kinds of things they buy. But an advertiser wants even more: are they listening to that song and then buying my products?
Targetspot has audio ad tracking and it is driven by view thrus not click thrus. People don’t click on audio (or video) ads. But they do recall with sufficient frequency and do transact. So you need to cookie and watch over a 30 day period
Do you then foresee a time when radio or Rhapsody like services are the new ‘record company’ and they take on the task of producing music to sell ads against it? Why pay the middle man when you can collect revenue on licensing and ads?
I think music-related sponsorship is another manifestation of that. Make a direct deal with bands/artists to either use their already-recorded music to sell products or have the bands/artists create music specifically to sell product. I think using corporate money to directly fund music creation makes sense.
It makes total sense in an era where album sales are declining but music listens/attention is up. While people might not pay to own music, they’ll certainly pay with their attention to hear it and that paradigm is a marketers dream. A radio station can treat an artist like an employee, put him on salary, give him profit sharing, distribute the work and sell ads against it. This works even better when you’re the guys controlling the distribution. You can ensure that your guy gets heard which contributes to getting him over, which contributes to charging more for ads.Bam, I just fixed the music industry. Where’s my cookie? 😀
Unfortunately I think its more likely that the labels will be the new rhapsody because their on demand licenses are so damn expensive that third parties can’t make money with subscription on demand services
Can you imagine that service though? Each song will cost you $5 for the first minute of music and $2 for each additional minute, it’ll be worse than a psychic friends hotline.
I’m trying hard not to imagine it
Agree Fred. Artists have to think of themselves as media companies to monetize the audience they’ve built. The different levels of affinity expressed by consumers in various social networks is an excellent means to quantify that media power. Our company, Colligent.com, does that for all the major record labels and management companies.
I’m starting to think the problem with ads is that people don’t have enough extra cash to buy vanity items. No industry optimization of ads will put dollars in Americans pockets unless we invent and implement them.Agree on the listens statistic and datamining though.
I agree that we must build smarter ad supported music sites that give back more money to the artists. And if so, Soundscan would build the proper tracking systems and pay attention. But I don’t agree that listens (streams) are as important as scans because listens is more like radio.Radio play doesn’t count toward scans. When people tell me that a song has been listened to on Myspace or Pandora 1M times, I say well how much of that converted into tangible sales or downloads?It’s much more easily trackable to account for sales of physical goods and downloads. Each good has a UPC or in the case of digital, an ISRC code.As long as Soundexchange keeps monitoring the # of online plays and giving money back to the artists and labels it should be status quo. A scan is a sale.
until people stop buying music and start subscribing to it. apple’s purchase of lala is an indication of where things are headed
Agreed, I think there is a mixed model business.
I completely agree. 5+ years ago the number of DVD’s you owned was a status symbol. Now, I can’t think of the last time I bought a movie.Ubiquity is the new scarcity, and Apple (of course) already knows this.In the immediate future however I’m interested to see how 4G, mifi, and companies like Pandora, LaLa, etc… revolutionize how we consume music in our cars (and tangentially for anyone in the music business, is there any correlation between car company adoption of technology and marketplace success similar to the old story about the Porn industry adopting VHS being the final straw in their battle with Beta?).If I were Sirius radio, I’d be very worried.
personally i’d like pandora and lala revolutionize my purchasing habits first.it’s quite a pain to buy a track i just listened to on pandora or last.fm. lala has come very close to how i’d like to be able purchase music–click a button and it adds that to my library. perhaps with apple’s integration of lala with itunes, i won’t even need to open my itunes browser anymore to purchase something. click a button, verify my account, and bam it’s done.and with these services having such a wealth of data about my listening habits, you would hope they leverage it to the best of their abilities to upsell (or cross sell?) music i might like.btw-you’re right. eff sirius. or radio for that matter. i think with the expansion of how we listen to music people are going to want the songs *they* want, not what some radio playlist offers them.
Once that conversion happens, and people start subscribing to all songs rather than paying for individual songs, do individual artists make more or less money and is the overall music industry bigger or smaller from a revenue perspective, compared to the golden age (80s, 90s)? I still think on a per artist basis, there will be less money, and the overall industry will still be dramatically smaller from a revenue perspective.
I still can’t quite grasp how you can monetize on attention without becoming an intrusion to consumers. Subscription is a good model because it takes away the hassle of having to purchase each time you want to listen to something you haven’t already own a copy of. So to customers, you’re essentially monetizing on giving them the convenience. You could be splitting the cashflow with musicians based on how many times each of their music was listened to. But still, it’s not really monetizing on attention.
Taking this on a parrallel, but there is a camp that believes that all media should move from actual planning targets to actual engagements. Why buy TV based on GRPs? How many of those people actually watched the 30 second commercial that you bought?The music industry measurement model (for record sales) is not very unlike the measurement of media. It’s evolving but not as rapid as many people would like.ErikSchwartz rightly points out below, “be careful what you measure, you might just optimize for it.” Seeing how that’s working out in media firsthand, it opens up unexpected doors
Look at the crazy programming shit radio does to try to optimize for Arbitron ratings (both diary and more recently PPM). They make programming decisions based on optimizing their product to the measurement methodology. Stuff that’s hurts the efficacy of their customers ads (and pisses off listeners) but looks good to Arbitron so they can up their ad rates.
The trip has only just begun:Reading “The future of the music album? Check out RjDj’s Little Boots app” on Ensembli: http://ensembli.com/stories…
music games like RJDJ are another example of engagement being the metric not the transaction
Listens is a valuable metric. But there are so many more around now like torrent seeds, tweets, blogs, etc that give a much richer understanding of what people are listening to. Billboard or Nielsen should step in and own this space.While artists like the Black Eyes Peas, Lady Ga Ga and Muse were huge this year, to me it is far more interesting that acts like The XX, Phoenix, Kid Cudi, Yeasayer and Passion Pit performed so strongly.stephen @ wearehunted
I would not hold my breath that the traditional measurement companies are going anywhere near this space. And if they do, the DNA in those organizations is not a good fit for radical reform.
I think they just might this time. Streaming is a new game. Punters are putting their money on the table. If I sell music monthly subscriptions, what stops churn? Data is gold. The artists want it. The labels need it. The tech is not rocket science. The market has moved whether they like it or not.
can i get an amen?erik, i think your spot on. imo, the companies that are able to iterate early and often (thanks fred, haha) like last.fm and pandora will realize the gold mine they are sitting on and use that for new services.at least, i hope they go that direction. e.g., not just monetizing via advertising. adding marketing data options, etc.
Totally off topic question. Has anyone put together a twitter list of the regular commenters on AVC?
shana carp has been working on this and is running into some problems
At first glance I think TweepML could help with this
Here’s Shana’s list – http://twitter.com/shanacar…
Been extremely busy yesterday
dang…what does it mean if we’re not on the list 😛
Nothing, there have been multiple requests to crowdsource the list.
Hi, umm the problem is how to develop a non-totally arbitrary list. Right now I try for people above 30 comments: One of my major issues is people coming and going from the community at large (there are clearly people here with large posting histories that come and go, that I may or may not know) I was hoping for a non-totally arbitrary cutoff of a month, but since there is only me doing this, with my own life and stuff circling around me, this is actually a problem.I wish I knew how to develop a very fluid twitter list to compensate, but…. It is definitely one of the more expansive notions of what a twitter list can do (or should do).I also wish I knew everyone very well, both online and in person (that would make life easier)….The only way to be definite and fluid about the experience, and be very fair (I’m sorry I’m a human, and have lots of arbitrary problems about me) would be to have a database of some sort of the posting history of the entire blog and do all sorts of funky things to it, and then develop a program that gave into both roving arbitrary cutoff dates and some sort of numerical posting history in tandem. I don’t have that data. And no, I don’t think asking for that data is going to help. That’s people private email addresses and ip logs. I’m not the person who should be having them, it’s none of my business, unless they choose to give it to me.Crowdsourcing is a way around this to a degree, but that can promote cliquishness. I’m not a fan of cliquishness.Either way, it’s a hobby project, and I started it because I had to keep track of people, so there you go. I’m guessing you feel the same need at this point.
So what disqus really needs to do is create a tool for making a rule set where a twitter list could be dynamically created and curated based on disqus behavior. (Now we’re fully off topic)
They’re thinking about it, from what I’ve been told, but that doesn’t help me. Nor does it help dynamic lists 🙂
the trend sure looks bleak: http://www.google.com/trend… but looks like it is still doing well in nashville, tn (maybe country relies more on these results).
This post is making me rethink something. I sold my shares in a tiny tech company that has some interesting technology for transferring music online. That company’s focus has been on facilitating the transfer of music from record labels to radio stations, but presumably their tech (which includes watermarking, tracking, and limitats on forwarding) could be modified to deal with end-user issues. Not that they’re event thinking about that, I’m just speculating. Food for thought though.
you are right on with this fred… we need both #s. I also am sure some of the copyright tracking patents from most of the big giants CTO’s that I have been seeing this past year are going to be focusing on this and reuse/aggregated partial usage…etc.
If I was Billboard.com I would be looking at measuring social and streaming not just digital downloads. Looking at Lastfm, Spotify, Rhapsody, Hypem and others and putting some good data together.
I have a great deal of ambivalence regarding this subject. Having worked in A&R, management, and licensing for years with artists on Sony, Warner and indies, I feel uniquely tortured/inspired by this conversation. My career path into technology was in large part inspired by the destruction of the record business.The ad model doesn’t generate enough revenue to support the industry, nor will it ever do so.Music is content that still needs to be paid for by the consumer because phonographic recordings cost money to create (this revenue is mutually exclusive to publishing revenue). If advertisers start subsidizing this, the only music that will “make it” will be music palatable enough for an advertiser– bad news.The best career paths I’ve been a part of were the instances when an artist I managed generated an excess of 80,000 music downloads (purchases) on their own- with a song recorded in their living room. I can say without question that this sales metric was exponentially more relevant than streams, which were well over one million. Why? He could pay his rent for a year on those revenues. Let’s do some math:80,000 singles sold on iTunes. $.70 net revenue on each, $56,000.001,000,000 streams online: let’s say the artist received a $20 CPM streams (which would be a miracle for an indie artist). 1,000 * 20 = $20,00020k is not enough to produce a full length record, go on tour, or pay for your band to tour for more than a month.If anyone has a solution to this, I’d love to start a new company around it.
christian, it’s great to have some numbers like this for our discussion. please share more if you can!regarding your comment, i think we also need to keep the consumer in mind:1) i hate being punished by licensing agreements. take pandora: i’ve fast-forwarded too many songs? i’m out of luck and need to start a new station. want to listen to that great new song again? nope, i can’t rewind. so now i have to go to imeem to listen on demand. oh wait, myspace bought it and jammed it into their crappy service. well, let’s go to last.fm. nope, they don’t have that song in their catalogue. for a business trying so hard to survive, isn’t this about the time where we as consumers should not have to put up with such a run around?2) it’s the internet, why is it also so difficult to buy the new music to which i’m listening? adding something to my music library should be as *painless* as possible. we are definitely not there yet. i would love to use pandora like this: if i want to buy a song, i click “purchase song”, verify my account, and click “buy”. and right there, without leaving the site, breaking my listening experience, or having to open my itunes browser (or the amazon store), those parties know they have a song to add to my library. maybe that is the technology you could get behind. i have a feeling that’s where apple is headed with the purchase of lala.btw, i’m definitely not a music industry veteran (unless you count being a music buyer/seller at a local store during my college years), so i look forward to your comment 🙂
I couldn’t agree more with your frustrations as a consumer, as it sounds like we both have the exact same experiences, and have tried the same products to circumvent this.1. Agreed on Pandora, and that licensing agreement is annoying. I will say though, I’ve met with the founder of Pandora a number of times and his philosophy on licensing is sound- even admirable. Pandora fought pretty hard to lobby against higher internet royalty rates for this reason (I do not recall what the previous streaming royalty rates were). If the rate becomes too high, the delicate ad model dies (because NO advertisers are paying $20 CPMs), unless you can negotiate lower royalty rates with a particular publisher. The major publishers were hoping to keep this rate high so they could cut deals with the Pandoras of the world. “We’ll chop your rate in half for our entire catalog”. What songs would get played more in that situation? The big artists. This would’ve crippled the democratic nature of a PandoraI guess , in conclusion, Pandora’s service suffers as a result of it’s license agreements, but I’d consider a great bridge to where we need to go.SELF CONTRADICTORY NOTE: royalties form internet streams are paltry– not enough to buy an egg sandwich for most independent artists. This is a bigger conversation but, i believe, an important one to explore. Artists can make millions from radio plays, but they make pennies from internet streams…2. I think the distribution problem will be settled when one player clearly dominates the streaming space, and I believe this market will right itself in the next few years. I bought one of the first MP3 players, an Arcos Jukebox, in 1998…..it held about 12 mp3s and the battery lasted about 30 minutes. We need some patience and the early adopters need to keep playing around with products until a winner emerges. Apple’s iTunes model was, again, a bridge into new media- not a sustainable model of music distribution. We’re already seeing tiered pricing emerge- which is a big step in creating an efficient market based on demand.Bottom line, people need to go back to buying music if they want to consume it whenever they want. Frankly, I think it’s bologna that consumers feel like they have some intrinsic “right” to consume that content on demand, for free. Artists don’t exist at the privilege of consumers, they exist at the privilege of customers. The idea that phonographic recordings are a loss leader for ticket sales is disheartening, and will devalue the quality of the recordings we get. Radiohead’s OK computer would not have been made if the objective was “make music to get on tour”. Recordings are not a commodity; they have value beyond the song.Sorry for the diatribe…
A few ideas:1. Bundles of tour/streams subscriptions (unless the share of people that buy tickets is fixed)2. Disrupt the labels overhead, Pandora can accept new artists directly (wonder what the labels reaction would be …)3. Put aside a part of the monthly subscription fee and let the customer decide to what artist it will go at the end of each month.4. Charge more for newer music. You can get a stream of hits from the last 3 months, but you will pay more.5. Find a bridge between buying a song forever and having it occasionally in a steam, a price point that will let you listen X times during the month/year
Great ideas. Challenges:A few ideas:1. Bundles of tour/streams subscriptions (unless the share of people that buy tickets is fixed)We actually did this with albums for one of our bands, and it wasn’t optimal (buy the album for access to the ticket). The idea was really savvy, but I think we lost a little on execution, even if it was hugely successful, it didn’t make up for the overall decline in sales– the hemorrhaging that the industry has experienced over the last 8 years is….overwhelming.Also, the subscription/ownership debate has been going on for a long time– i don’t think subscriptions are the answer. People need to want to own music, because it inspires them and makes them feel something. When they buy it- they need to own it for any purpose they want it for. I don’t think people rent their passions– and music, at its foundation, is a product of passion.2. Disrupt the labels overhead, Pandora can accept new artists directly (wonder what the labels reaction would be …)Pandora does this now– they can add any artist they want, they just pay a compulsory royalty for internet streaming.3. Put aside a part of the monthly subscription fee and let the customer decide to what artist it will go at the end of each month.See my thoughts on subscriptions- i DO love the idea of assigning money to the artist of your choosing. Maybe this is an internet streaming radio for $10/month. A the end of the month, you get to “buy” any album you want, based on the music you listened to all month?4. Charge more for newer music. You can get a stream of hits from the last 3 months, but you will pay more.I like that– but again, I didnthave to pay to listen to radio, so why will i pay now if I can’t own it?5. Find a bridge between buying a song forever and having it occasionally in a steam, a price point that will let you listen X times during the month/yearThe long-term lease is good! If I had that when I was a teenager, I would gladly give back all my old LL Cool J records 🙂
Thanks, you bring really important insights. Many things about the new business models has yet to be solved, and its crucial to have that insight. It’s a re-drawing the old demand/supply curves. The brave new world should, after all, feed the non mega-stars content providers at the end of day
i remember having quite a few conversations with reps from the publishers about the new consumer mindset, e.g., consumers seem to feel that if they buy the music (mp3, cd, whatever) it is *theirs*. the reps would point out: it isn’t completely the consumer’s property, they don’t own the masters, they can’t do all and everything they want w/ the music. this mindset frustrated them because consumers were being trained to get what they want for free.i am almost sold on the opinion that the free-sumer isn’t helpful, but i still can’t break from fred’s idea that listens are the best metric as we move towards streaming. why? because it’s a more nuanced metric than sales. i love music. i buy a lot of it. i can’t tell you how disappointed i am with some of the albums in my library. i wish i could give some of it back for a refund.think of it this way: lots of sales *and* listens? you’re a star. decent sales but almost no listens? your audience must not be the internet-type. almost no sales but a lot of listens? maybe you need to rethink the artistic direction of your next album (like listening to your producer this time around).i think there’s a wealth of new mindsets available in who listens to your music, how often, what else do they listen to, and where are they located. for instance, i think that a disparity between sales and listens is a worthy point of departure for investigation. but maybe i’m biased because my job description includes managing analytics and reporting for our clients’ websites…anywho, i appreciate your diatribe, and reminding us that artists need to make a living too. but everyone keeps playing hot potato with the cost of transacting music and i want to figure this out as quickly as possible because i’m tired of living in the in-between stage of not wanting to pay $10 for an album where more than half the tracks are mediocre only to put them on a hard drive that is almost out of room, versus the cloud-hosted services like imeem was where i can listen to a few singles over and over again.i’ll end on a sidenote: did you use muxtape before the labels/publishers threatened the site into oblivion? it was the absolute best music experience i had online, ever. and it had a ridiculous amount of potential. sigh…it exists now in a place for bands, but unfortunately hasn’t (i feel) really caught on.
i agree with all of your sentiments- especially that the album format is outdated.I used to use and loved Muxtape. It’s funny, because I feel like I’m defending the music industry in here- which I have zero interest in doing. The business needs to be restructured. And, its not that I think that we shouldn’t be measuring listens; I think it’s an indicator of an artist’s ability to succeed. However, if we can’t figure out a smart way to get artists paid for creating music, it won’t matter what me measure.
i forgot to ask…how is it that radio plays can generate millions? i am not familiar with that system, and it’d be great to know how it works 🙂
Thanks for the thoughts– I really love this conversation. I’m hopeful for the future and, like you- I just want something that works– and I want great artists to make great records.Radio royalty rates are higher than internet streaming rates. Also, because music is consolidated on terrestrial radio, a very small number of artists make big bucks- and no one else gets any play. It’s an all-or-nothing game, where the internet is more like half-a-penny for everyone.
christian, i’ve really enjoyed his conversation too. i’ve been traveling for the holidays, so it’s been difficult for me to continue it. in the mean time, i was wondering why in particular are radio rate higher; by consolidated, you mean the most number of listeners are there? if that’s the case, when/how do you think it will change?
so the theory that increased tours revenues are the payoff for low stream revenue is false?
i’d like to know the answer to this as well. i don’t mean to come of defensive, it’s just that i’m super-curious about the numbers behind this.for instance, say someone created a soundscan for “listens.” this included data from pandora, spotify, rhapsody, last.fm, and others. artists get data on the geographic location of streaming, demographic and listening habits of the users, and other such info. wouldn’t that be valuable data? so a band/manager could better gauge which cities to spend marketing dollars, tour, and what kind of merchandise to bring?
This is absolutely a valid point, and not defensive. The way touring works (for a lot of artists), is that they tour on guarantees from promoters. Promoters offer an artist a g”guarantee” in exchange for the artist performing, and that guarantee is a function of a promoter’s best estimation of what they think they can get in ticket sales .The issue: there’s no empirical evidence to prove that increased online streams is a great indicator of increased concert tickets, or increased willingness to pay for tickets. You have to keep in mind, these people aren’t paying for anything when they stream– they’re generally less likely to pay for anything else related to that artist.Put another way, there’s no way I would base my band’s routing off of how many people hit the “Demand It” button on their Myspace page. I would definitely base it off of transactional data (single sales, merch sales, etc). I don’t care about free-sumers because they haven’t really supported the act yet, financially– so what makes me think they’d buy a concert ticket?
The idea that tour revenues have increased is false. The idea that these revenues are becoming the lion’s share of an artist’s revenue is true.
its surprising. i thought that the “on tour” alerts actually increase tour revenues. Maybe the alert should appear before the tour is planed and closed. This way increased attention may actually result in increased demand, and the tour accommodate for that?
The number of listens makes for an interesting stat, but I don’t see how you can directly link it to artist compensation. It’s just too easily gamed.However, I agree that having the technology to measure scans, listens and their respective geography does enable new possibilities – including early identification of music hits. For example, if you have a relatively high listen:scan ratio (for a “low” number of localized scans) that seems to indicate you may have an emerging cult following.There are a new category of music startups that are helping bands and labels to source and analyze these kind of trends (e.g., Band Metrics, Music Metric).(By the way, on a completely separate note – Music Ally posted their A-Z of music startups yesterday, in case you missed it: http://musically.com/blog/2…
I can’t believe the music labels are still using scans!
http://www.nytimes.com/2009…Looks like the Arbitron is on it…
PPM is a joke for a lot of reasons. Here are just two (there are many more). First, you have to care enough about radio to agree to be on a panel. This skews the results towards increased market penetration. Second you have to be willing to wear something that looks like a pager from 1990.Is PPM better than the paper diaries still used in most markets? I suppose so. But it’s taken them 7-8 to roll out PPM so they’re not a nimble bunch so don’t count on them to move fast.Smartphone PPM metering would be more interesting. You couldn’t do it on an iPhone (no background apps) but on Android you could.
I’ll always be a fan of radio. While admittedly they are not nimble bunch, they’ve got that vast glorious built in market of people driving in their cars.Like an app, I’d think what they’d want it some kind of dedicated ambient listening device that was looped 24/7 back into a cellular network, so they could actually get real time reading off their sample size.
I couldn’t agree more. It is time we start paying people for their work again. And that we also think about the people behind the scenes. My brother works for a publishing company doing administration of royalties. There is an attitude that it’s only the artists and big time execs that make money so it doesn’t hurt to grab songs for free. But the guys like my brother who work in the music biz not for money but for the love of the music are getting hurt just as bad.
What you are proposing would be incredibly valuable from an ‘music impression’ standpoint if the ad deal were built on a sliding scale.What Soundscan brought at the time was a form of legitimacy to the sales side, both from a Chart and an artist royalty perspective.With the speed at which todays music is hot/not creating a new Chart, ie “Today’s Most Played Song,” encompassing streams, scrobbles, sales and radio, would be more valuable to an artist than placement on an old magazine chart.
Fred – you are describing exactly what NextBigSound (one of the TechStars 2009 companies) is going after – http://nextbigsound.com/. Here’s the Lady Gaga example – http://nextbigsound.com/nbs…
Fred’s call-to-action above is ‘let’s start measuring it correctly,’ and I’m not sure that it’s not being measured correctly, but perhaps not as publicly as via services like comScore/Compete.I work with all four major labels, and meet with them for updates. I am shown detailed measurements of how (i) radio, (ii) YouTube views, (iii) streams and other inputs tracks in alignment with transactions (in my case, of ringtone sales). I see pie charts where revenue is sliced into more than a dozen tranches. I see rigor where the dollars matter.At the indie level, firms like ReverbNation are providing hosted ‘Music CRM’ that to me approaches parity with Salesforce.com. Indies can measure an incredible amount of data, although I’m sure managing a high number of logins has become cumbersome — there undoubtedly will be efforts to further-consolidate online/offline consumption data.So for those whose business it is to measure all the forms of music that happen, I think measurement happens. Whether an aggregator like comScore keeps up seems to have played out already. Put another way, does anyone’s internal web metrics foot with what third-party services say? I know that apples-to-apples comparison is the value, but for whom?
for many years in the music industry it was fashionable to say “how do we compete with free?” “tough to compete with free’ etc etc . . . i think the real question is “how do we profit from free” and in a true way what I am really saying is: how do we monetize attention. thanks for another great post. fun times ahead… have you looked at the Guvera model?
Interesting post Fred (sorry I’m a few days late here). I think listens is a metric that is closer to actual consumer activity, but I’m not sure it is the right one to replace scans with.Labels still get paid on “ships” (shipments of CDs to the stores, even though they can be returned back to the label/distributor). I’ve had many people from major labels tell me that units shipped (not units actually sold through to consumers) is one of the metrics that the very top level of management (or corporate parent in some cases) focuses on, because it is the gross revenue number, and therefore labels focus on getting as many units out the door as possible. Even though many people doing the day-to-day at the labels know shipments is a mostly worthless number, it is still the one that performance is tied to. For those who aren’t aware, the RIAA awards for things like Gold, Platinum, and Diamond “sales” are actually based on shipments, not actual consumer sales/scans (see: http://en.wikipedia.org/wik….That said, scans are still focused on, because for physical product, it is the only indicator of how many of the units shipped sell through (and aren’t going to be returned), but as you point out, the metric has not evolved much.We’ve been thinking about this a lot, and I believe the new key metric should be # of Engaged Fans (or a slight variation on that). Plays (from last.fm, MySpace, or otherwise) are an input to the algorithm to determine engaged fans, along with other fan interactions with an artist (pageviews, opens, clicks, sales, etc), but I don’t believe looking at plays is a fully accurate picture as it is just one lens of the value of an artist.
Thanks Ryan, I wrote a post last week on the NBS blog about how consumers take totally different actions now when they first hear a new band beyond calling into their radio station and picking up the physical album at their local store (both of which are measured by (SoundScan/BDS). The full loop that needs to be measured IMHO:1. Artists take any number of career building activities (i.e. interviews, concert, new release)2. People who are exposed to those activities take any number of actions that can be measured on a more granular level than was ever possible before (i.e. friending on a social network, streaming the music, commenting etc.). We are squarely focused on measuring and visualizing this behavior.3. We can overlay revenue numbers to see which spikes in social music activities actually translate to dollars in the bank and thus, which career building activities were worthwhile.Glad to see this conversation taking place on AVC!Alex WhiteCEO at Next Big Sound
“3. We can overlay revenue numbers to see which spikes in social music activities actually translate to dollars in the bank and thus, which career building activities were worthwhile.”Or even perhaps just which social music activities lead to higher CTRs. The end result being the same kind of detailed analytics that you can get from Mixpanel or Google Analytics, except for music activities instead of webpage activities, or music activity embedded in webpage activity.Some of the best value I’ve gotten from analytics products has been through the customizability offered by APIs, making it easy for me to create custom reports, etc. I’m curious about what the differences would be in a music-specific analytics API, and whether it be possible to incorporate A/B testing or other forms of hypothesis testing.