With the launch of Apple’s new streaming service, we’ve seen a resurgence in the popular arguments over streaming: Are the royalty rates too low? Can this possibly be sustainable? Does streaming devalue music? If you can hear everything for free, why would anybody buy anything?
For the armchair pundits pompously pontificating in pubs, the answers to these questions don’t really matter: they don’t have any skin in the game. They’ll keep doing their thing.
If you run a label or release your own recorded music, though, the answers to these questions do matter, because at some point you’re going to base a decision on them.
I spend a lot of time with labels, and the decision many of them are wrestling with is this: “Do we stream our products or not?”
This is the wrong question to ask.
The record business is old. Edison patented the phonograph cylinder in 1878, and discs were introduced to the US market in 1889. Since then, we’ve had 78s, 45s, LPs, 8-tracks, tapes, cassettes, CDs, SACDs, DVDAs, Pure Audio Blu-Rays, MP3s, MP4s and FLACs.  People act like downloading represented a big shift in music consumption, but it really didn’t. 
Except in few rare cases, the shift to purchasing digital downloads hasn’t seriously challenged our idea of what constitutes a recorded music product. Digital albums are, for the most part, simply digital representations of the CDs they either duplicate or replace. Each format strikes a different balance between convenience, quality, playing time, and durability, but what they all have in common is they are all sold.
The business we have has evolved—partly through intelligent planning, partly through natural selection—around the process of convincing customers to part with money in return for the permanent ownership of recordings. 
You make a record, promote it to people who have never heard it, and get them to pay a chunk of money in order to be able to listen to it at any time in the future. Thus it has always been.
This is important, because our entire experience of recorded music, all the assumptions we make about what constitutes a product and how it should be valued, have been shaped by an ownership-based market that is more than a century old. Like unmetered water, an all-you-can-eat buffet, or an unlimited cellphone contract, the new streaming music services create a different set of incentives for both customers and suppliers. To apply these changes to a mature ownership-based market, we have to forget a lot of things we didn’t even realize we had learned.
The question we have to ask is not “Should I stream my stuff?” but “How does the existence of streaming services change my job?”.
It’s perhaps helpful here to look at what the movie industry does. Rental (and even subscription) has been a big part of their business for decades, so they’ll have had the chance to learn from their mistakes as the market matures. You don’t have to hang out with a Hollywood lawyer for long to realize that while the dumbest people in the movie industry are just as dumb as the dumbest people in the music industry, the only ones who take on the major labels and win are the studios. The men in suits might know a thing or two.
Blockbuster movies these days are so homogenous they make the Top 40 look like a hotbed of artistic rebellion, and yet studios are quite comfortable distributing them through a wide range of outlets. There are previews followed by a wide theatrical release, traditional physical rental, home-delivery rental, cable pay-per-view, download-to-own, digital rental, online subscription services like Amazon Prime, DVD, Blu-Ray, in-flight entertainment, cable movie channels, and network broadcasts. If there’s a way to get paid, the studios are all over it.
Indeed, for as fun as it is to lampoon them for making the same movie over and over, we should look in the mirror once in a while. The economics of the physical sales model have got us thinking an album has to be at least ten songs or 50-70 minutes of music. When iTunes came along, labels and artists alike bent over backwards to try to keep the album format alive instead of realizing what an absurd creative and commercial straightjacket it has always been. This is particularly stifling in the new music world, where a new work might not see release until there’s a full CD-length program to keep it company.
Studios know that people have appetites for 15-minute cartoons, 43-minute episodes, 120-minute feature films, and 27-hour stimulant-fueled Breaking Bad binges. Movies are profitably made for straight-to-DVD release, and video streaming services are making their own content and dropping whole seasons in one day. The content might all look the same, but the business plans are sophisticated, carefully tailored to the content, and different.
Other businesses have embraced (or been forced to accept) access as an alternative to ownership. Public libraries did not destroy the book business, despite its long-touted decline.  It took quite a while for recordings to replace sheet music as the most visible form of music consumption. (The first Top Ten charted the sales of sheet music, not records, and sheet music first became available on subscription 250 years ago.)
Of course it’s possible that subscription streaming will be the end of the record business as we know it, but I’ve never understood what is supposed to be so great about the record business as we know it. Let’s worry about something that is within our power. What are we going to do about all this?
The way I see it, you have some action items:
1. Stop looking for a new model. When people say they’re looking for a new model, what they really mean is “can somebody please come up with something that works so I can copy it.” It is going to be harder than that, but only a little bit.
2. Stop thinking of your recorded music business as “selling albums.” From now on, you commercially exploit the copyright in audio recordings. Write yourself a list of all the ways you might do this. Include not just paid streaming and downloads, but licensing, future compilations, and free downloads and streams used for promotion. Consider them all for every recorded asset. Try to keep an open mind about what constitutes a product. Do not wait until you have a whole album to think about this. Alongside your marketing plan, make a release plan for each product. Do not simply take the last one and change the album title and the date.
3. Make worse records. When I buy an album, I’m expecting a certain baseline level of quality, because while only a fraction of the cost goes toward creating the content, I’m still paying a lot of money to permanently own that recording. On a streaming service, I already paid somebody else. All I invest in your product is the time to listen to it. I’m more willing to take a chance, and less likely to be disappointed. There are no refunds for bad records on Spotify, so many interesting-but-not-sonically-perfect live albums have a place on streaming services even when they really don’t belong on download stores.
4. Make better records. To make money from selling albums, you have to convince people that they may, in theory, wish to listen to them at some point in the future. To make money from streaming music, people have to actually listen to it. If your albums are better in theory than in practice, streaming platforms are not going to be the place for you. You have to record something that doesn’t already exist, and which people will want to hear. If you want streaming subscribers to buy an album you’re not prepared to stream, then it really has to stand out to people who haven’t bought it and don’t go to record stores. It will cost the same as a month of listening to everything on Spotify.
5. Remember that “streaming” is not a single service with a single deal. Alexander Street Press and Naxos both offer academically focused streaming products that are priced higher than Spotify or Apple Music, and pay correspondingly higher per-stream rates. These services already offer a happy medium for labels reluctant to participate with Spotify and Apple Music.
6. Consider windowing. Apple Music will let you set a streaming release date that is some time after your download-to-own release date. There are ways to do the same with Spotify. Your commercial goals may be best served by not releasing the DVD the day your movie opens in cinemas.
7. Consider why you made the recording in the first place. Not everybody’s primary motivation is profit. If the priority is to reach a large audience, and to get people to take a chance on your music, streaming might come quite high up the release plan.
8. Try stuff. Innovation is just having something to show for playing around. The more fundamental the threat to your business, the more important it is for you to play around. Most big companies are bad at this, which is why they’re so often late to the party, and when making a serious plan to invent something doesn’t pan out, they use their money to get what they want instead.
9. Build a following. Streaming service providers are determined to turn their jukebox apps into social networks. It feels desperately contrived, but it is happening, and it isn’t enough to get people to like your album once. For you to get paid, people have to listen to your records over and over again, and it is nobody else’s job to make that happen. If you’re not streaming, this still matters, because if a major discovery platform doesn’t have your music, you have to work harder to keep the same level of visibility.
10. Don’t forget your back catalog. A year after release, you might have shipped 80% of all the albums you’re ever going to sell, but on streaming services, the work is just beginning. Make sure your catalog is nicely linked up online, that any resources about the music have links to the recordings, and that you’re using playlists, editorial, and any other tools at your disposal to get people from one of your recordings to another. Make sure the metadata is lovely. Put the sleeve notes on your website—somewhere obvious. Look after your old records, and they’ll continue to look after you.
If you’re in the business of making and selling records, then streaming means your job has changed, and it’s not as simple as opting in or opting out. Whether you want to stream or not, things are different now. The one thing you mustn’t do is ignore it. Good luck.
a) It is unsustainable. The argument goes that streaming services are not profitable despite their huge popularity, and therefore never will be. This misunderstands the nature of investment and the projected growth of these businesses.
b) It is a scam. If the majors are screwing their artists, then this is (i) not new and (ii) between them and their artists. It is not an inherent flaw in the delivery mechanism, nor is it Spotify’s fault that some people entrusted notoriously devious multinational companies with the exploitation of their intellectual property.
c) It is not transparent. This is not true either. Streaming services pay out a fixed percentage of total subscriber revenue according to each rightsholder’s share of the total number of streams. The formula is not complicated, although the implications of this are not always obvious.
d) They insist that every stream is worth the same, and that amount is too low. This is the only argument that holds water, and it is a straightforward business decision: the service offers to pay you X each time somebody listens to your music. Take it or leave it. This isn’t a moral question, it isn’t about transparency or power or big guys and little guys or the contract to take photos of Taylor Swift on tour. It’s your music; they are offering to pay you for it. You decide.
Some products will reap poor financial returns on streaming services because they are, by their nature, not something people listen to often or repeatedly. Some labels have catalogues comprised entirely of these products. If those labels intend to continue making exactly the same products without regard for the changing shape of the music market, they would be well advised to steer clear of streaming services altogether, but that does not mean they will not feel the effects of them.
2. At this point, even downloads have been around for a long time: I run a label for King’s College Choir. The choir itself has been around for half a millennium, but fewer than half the singers are older than the iTunes Store.
3. Outside of chart pop, which has suffered badly from a sudden increase in ways for young people to express their individuality (or lack thereof), the big shift came a little earlier with the widespread success of online CD sales. This has had a huge impact on the diversity of available recordings. iTunes and Spotify also have almost everything**, but they didn’t start this. Amazon did.
**There are exceptions. Garth Brooks isn’t even on download stores.
4. Usually round ones, as Will.I.Am observed in a moment of either inane idiocy or surreal genius.
5. There’s an old joke that the first book published using moveable type was the Gutenburg bible and the second was a book about the death of the publishing industry. With some regrettable exceptions (I’m thinking Twilight), taking publishing out of the hands of monks with nice handwriting turned out not to be such a bad idea.