I would happily pay a significant amount of money every month (say about £30-40) for an all-you-can-eat buffet of DRM-free downloads covering all the released music (since the beginning of recorded music) by all the major labels.

This would translate into some £480 of income from me for the recording industry, versus £0 at the moment. I'm pretty sure I'm not alone. I wonder how much money they're leaving on the table by offering inadequate products like Spotify.

Spotify does more than let you listen to the radio - they have a large catalogue of music that you can search through and play at whim.

You can also save music to your mobile device and listen to it later. If you want, you can go to a friend's house, login to Spotify on your account and then listen to your music there.

http://www.spotify.com/uk/about/features/

All this for just £10 a month! For all intents and purposes, Spotify is "all-you-can-eat" for the needs of most people. I wish they'd improve the range of tracks they have but that's somewhat beyond their control - this will always happen until the market tips enough to get certain old-world businesses on board.

Before this comment runs the risk of sounding like a hagiography, I should point out that the features for creating and sharing playlists are fairly limited and the client software is fairly basic (at least on OSX, which is the only version I've used).

Edit: Beaten to the punch by ladon86 by a couple of minutes. I should learn to type faster, or perhaps just use fewer words :-)

I don't care that I'm left with nothing when they fail or change their policy so it doesn't suit me and I decide to take my money elsewhere. They offer me a convenient service. I'm not even interested in buying CD's.

Spotify isn't a radio service, it does have a (fairly iffy) recommendation service, but it's a fundamentally a "I want to hear this track or album by this artist, there it is, click play" service.

Think about it as the equivalent of buying a measly 3 or 4 CDs a month back in pre-internet days. I don't think that's an excessive amount of music if you're a music-lover.

I would easily download much more than that each month, and would be happy to pay for an omnivore plan to match.

Do you not see a risk to the company offering an unlimited download package that some (many?) would subscribe, download everything they've ever heard of, and can think of, then cancel? Or after 2 years of subscribing if they announced they were changing their model that people would do the same?

Now that this is not the distribution model anymore, I am certainly not happy to pay £30 a month to get the equivalent of 3-4 CDs. Times have moved on. £30 a months should buy access to everything, not just to 3-4 CDs' worth of music.

90% of the price of CDs is distribution and packaging. Those costs are nil on the internet, and therefore the price should be adjusted accordingly (and the record labels go bankrupt). If they can evolve and start providing a better value-add service (like, maybe, a very good interface for finding and downloading music) then they might be worth something. Atm the labels are just a leech on the back of artists and the public.

Anyway, if you must sell music trac-by-track, I would consider a per-track model that costs 10% of the price currently being used - i.e. 10 cents per track. But I probably wouldn't spend as much as if I was offered an all-you-can-eat subscription.

The vast majority of cost of producing and selling a hit album is marketing, and that cost hasn't changed significantly for the labels, at least not yet.

It's quite possible that the economics of 10 cents per track might work out, but it'll be without the major labels marketing teams.

Whether that should bother anyone, that's the real question.

Not sure your model works.

Also, I'm uncertain that millions or hundreds of thousands of people really do pay £40per month for rapidshare/usenet accounts. I'd like to see some evidence for that claim.

This only makes sense if you're talking about someone with tragically limited tastes in music. There is no way you could ever listen to "all the music you want" in a lifetime, let alone one measly month.

There is more good new music produced and released in a month than you could listen to in that month, before you even consider the entire history of recorded music that already exists.

I know a few usenet/rapidshare service subscribers, they download more HD movies per month than they could watch if they sat there 24x7 watching them, never mind the amount of music they could download if they turned their attention to it.

And this is essentially a thought experiment - extend it out to a year, where you're downloading 24/7 and what I'm saying holds true. Paying a fixed amount for (legal) unlimited DRM free downloads isn't a model that's going to work.

And it doesn't make sense to consider someone downloading 24/7 because unless you're on dialup, you can download a lot more than 24 hours of music in 24 hours.

However it is clear than the value you get from spotify and similar services is very much determined by your attitudes to music. If you cast your net wide and look for gems from across all styles and niches of music, then it's easily worth it. If you just want to listen to indulge your relatively static tastes in the one or two areas of music that concern you, then its a less obvious choice.

The question is, are there enough obsessive people who are into Music in the wider sense, and have a need for on demand "random access" to any and all artists/genres, to sustain spotify? Because they are the people for whom the service makes sense. People who want to listen to the same 50-100 albums on rotation are unlikely to be persuaded by the pricing model.

PS. Yeah, "tragically limited" was a thoughtlessly flippant thing to say. Didn't mean to be so condescending. Compare it to the attitude of a hacker looking down on someone who doesn't feel the need to learn anything except Java because it "does the job" :)