See, as I wrote in Is it time for Tip Jar?, I’m all about looking for solutions in the face of the death throes of the music business. If I really respect music then my challenge today is:
What would it take to make streaming work for those who make the music that’s streamed?
Like any good thought-experiment, we have to limit our scope, and put some variables out of play. So, let’s assume:
- Profit isn’t the motivating factor
- There are listeners who understand what it means to say #IRespectMusic
- The bureaucracy of the music business – rights collection, etc. – are waived (voluntarily)
- When I say musicians, I’m including the producers, song-writers, etc. who contribute to making the music
So, ladies and gentlemen, our aim here is to imagine a streaming service that works for the musician AND for those listening
[we should definitely all try to use many, many more YES… AND… statements instead of NO… BUT…]
In broad terms, there are two potential payment methods:
- Subscription – pay one amount, listen as much or as little as you want – fixed cost to listener [GOOD], non-scaling revenue for musicians [BAD]
- Usage – Per play – variable cost to listener [NOT-SO-GOOD but at least FAIR], scaling revenue for musicians [GOOD]
So, given that this is an AND solution, we’ll go with a Usage model
[one of the few times most musicians will accept a Pay-to-Play approach, methinks]
Because we’re entrepreneurs, let’s give our streaming service a snappy name
[not to be found anywhere in the English language, of course]
[see what I did there? Though I do reserve the right not to consider whether listeners and musicians to Music Fayre would be known as MusicFairies…]
There are some parameters we need to set for the sake of the experiment:
- A song unit is 3 minutes long – 20 songs an hour
- A monthly billing period is 4 weeks long (28 days, without the fast-moving zombies)
- Every stream counts, partial or complete, long or short
[trust me, it gets very complicated very quickly if we try and work an algorithm]
Let’s look at a couple of use cases and see where we end up.
Case 1: Bob
[actually, his name is Robert Paulson, though I’m categorically NOT talking about that]
… anyway… Bob is a pretty hardcore listener. On average, he listens to MusicFayre for 5 hours every day of every week:
- 1 hour = 20 songs
- 5 hours = 100 songs
- 28 days @ 100 songs per day = 2800 songs
So, Bob listens to 2800 songs.
Case 2: Tyler
[still NOT talking about it]
Now Tyler, on the other hand, only grabs an hour’s worth of MusicFayre on an average day:
- 1 hour = 20 songs
- 28 days @ 20 songs per day = 560 songs.
How much might we charge for a song, then?
Well, let’s say that a permanent download at iTunes is 99c – essentially an unlimited single-user license. That’s our upper ceiling, I guess. Though, according to this page
[which is on the internet, so must be true]
an artist gets 9c per song downloaded.
[that’s less than 10% for those without calculators]
Let’s call that our floor then.
Now, being a business, even if it’s one that has placed it’s profit motive on hold, let’s say that MusicFayre runs very, very efficiently and only has to take 20% margin to cover operating costs.
[it’s a thought experiment, remember!]
If we charge 99c per song:
- Bob pays $2772 per month, Tyler pays $554 per month
Nope – ain’t happening, even if the musicians do make 79c per song.
So, let’s look at the other end of the scale – 9c per song:
- Bob pays $252 per month, Tyler pays $50 per month
That’s high, even for Tyler – and remember, of that 9c, nearly 2c is going to operating MusicFayre, so musicians are making LESS than they would for a downloaded song at iTunes. With traditional radio free to listeners, and streaming services willing to go much cheaper, simply by starving musicians, the barrier-to-entry for MusicFayre seems just too high. There’s altruism, but it’s a big ask to believe our listeners are THAT altruistic. Bob, especially, is likely to raise a sardonic, cynical eyebrow at least…
You know what? I’m going to make this a series of posts, because this is already getting long
[warned you at the start, right? RIGHT?]
but let’s summarize where we’ve got to, because even with some pretty ballsy assumptions, the thought experiment tells me:
A streaming service that works for BOTH listener AND musician is unlikely to succeed based upon a listener usage model
See you in the next post…