« For those of you who didn't get Rock Band this Christmas | Main | The first Long tail Olympics »

January 03, 2008



Well, for what it's worth, Reznor notes that it was expensive to produce and distribute (bandwidth, costs of creating the site) the album. So even though nothing was spent on marketing, the studio costs may eat significantly (he doesn't give numbers for this) into the increased amount of revenue.

Also, Reznor isn't completely negative about the project as you seem to imply. He concludes the post with a positive description about Saul's increased visibility.

Dane Cao

the whole thing is impossible here in China!Nobody will ever pay to get the file with enhanced quality,cuz they know soon they will get it via amule or somewhere for free...I think this goes way beyond just cultural differences.


Thanks for this post, I just realised I hadn't downloaded this yet. I've downloaded the free version now, but if I like it I'll go back and pay.

As npdoty has said, studios are often hugely expensive. Even if, as Reznor probably has, you had your own recording studio, you'd need to pay your recording engineer and mastering engineer a substantial amount to have something of releaseable quality.

Also the visibility arguement is a double-edged sword. Fans are unlikely to pay for something they have previously been able to get for free. This is an arguement used by many music enthusiasts (and independent record companies) regarding the sale of mp3s, as they feel it will only encourage piracy.


Furthermore, 192k is not "medium quality" unless you're an audiophile purist. Most downloads (legal and illegal alike) are 128k. If they released a free 96k version and charged for a 192k version I bet they would have made a lot more.


As others have said, I think the issue of success or failure here relates to the scaling of the project.

A typical recording budget for a major label artist would be somewhere between $100K and $300K. If they spent that sort of money, or even say, $75K, recording this record, then I can see why they might be disappointed (especially if the bandwidth costs were high). Anyone care to hazard a guess about how much those might be on something like this?

Of course, very few major label releases recoup their recording budgets, especially in the frist 3-5 years after release. It's a complicated betting game. Most fail. But the hits subsidize all the failures, because these launch costs are very front loaded, and once something is a hit, the profit per unit goes way up.

If further experiments show that 15%-20% of the total number of people downloading a record are willing to pay $x for it, then that is a model you can at least begin to work with.

From there, the question becomes, how many people are likely to download it, and what steps can I take to increase the absolute number of people who will download it (be they paid promotional steps, adjustments to the price, or cheaper guerrilla/grassroots marketing steps)?

Armed with a little clearer picture about some of this stuff, it would be easier to make an educated guess about how much you ought to risk on the recording budget and promotion budget.

There is definitely nothing quite as good sounding as a record made by a great band and seasoned recording professionals, tracked onto 2 inch analog tape, with vintage gear in a cush studio.

But as between a $50k budget and a $250K budget, it usually doesn't pencil out from an 80/20 rule standpoint. Usually, you can make a really nice sounding record for $30k-$50k. The extra $200k gets spent in an effort to achieve that extra 20% of polish (that and people will certainly charge more for their services when they know the budget is higher).

To the extent that artists, labels, etc., abandon the quest for the blockbuster, data like TR's may help them to craft an appropriate model for making money selling music on-line. In addition, widely disseminating the music may also have collateral benefits from a promotional standpoint.

There's really nothing new in this. Indy artists and labels have understood this for years. The successful ones are almost always bottom up bootstrappers.

The ones that succeed often aren't the well capitalized, trustafarian labels. They're the ones where the folks involved did everything as cheaply as possible and only put more money in when circumstances indicated that it would have an actual impact.

Death Cab for Cutie, Modest Mouse, Blue Scholars, and a whole host of other successful indy artists didn't get big by off of some large cash infusion on the front end. They started small, and scaled their projects correctly as they grew, so they were more likely to break even (or make a profit).

This is why hip hop makes a lot more financial sense right now than a lot of indy rock. It's just much cheaper to make it. A creative hip hop producer can easily make a great sounding record for $3k-$5k, because it can all be done in the computer, it doesn't require a fancy acoustic environment to record it, and you don't need much in the way of mics or outboard gear.

Even if one spent another $25K on press/promotion for such a project (and adding in bandwidth costs), you'd still come away doing quite well if 28k people pay $5 a piece for such an album. Indeed, you'd probably break even around 10k sold.

As others have said many times already, for the unknown artist, the big issue is simply getting anyone to pay attention at all (let alone pay money).

Saul Williams was obviously already at a decided advantage. So the results probably skew high. He had a very high profile patron. His last record had sold more copies than probably 90-95% of all records released (I know that may seem like a crazy statement, but very few records sell more than 10,000 copies). In many respects, he is already above the 90th percentile of rock success.

But in the tournament of rock, where only the top one or two percent actually make Trent Reznor money, being in his position is a lot like being the starting power forward on a CBA team. You can credibly say that you are one of the 1500 best basketball players in the world. But until you are one of the 400 best basketball players in the world, you're not really going to make great money doing it.

Notwithstanding that, if one is content to consistently hit singles rather than prematurely swinging for the fences, I think there might just be some optimism to be found in this Saul Williams story.

It's just not going to be the obvious utopian fantasy of most musicians (myself included), where you make the recordings, put them up there, a respectable number of people listen to them, and each listener pays you $5.

Nope, it's going to be even more like phone sales than the existing model, with a lot of traveling salesman added in on the side of good measure.

But as someone said on the Reznor blog comments, maybe there's a way to release stuff like this via torrent, so that the artist doesn't bear the sole cost of the bandwidth. And maybe there are ways to help supporters promote your project through banner ads on their blogs and stuff like that.

Kudos to TR and SW for doing some more market research. At least it's something tangible people can react to and brainstorm about.

Petter Duvander

Ryan: I disagree, releasing lower quality would be going backwards, but I see absolutely no reason for Reznor/Williams paying for bandwidth to supply free downloads. Mr. Trent's used "official" bittorrents before, and I don't see a reason for that not to work in this case. A fan who, for some reason, doesn't want to pay for an album surely will be glad to at least share his/her bandwidth to promote the download.

Another question is whether it was clever to go for a fixed $5 for the full versions. Personally, I'd probably have payed a lot more for the album had I had the opportunity (your currency isn't what it once was). Downloading multiple times to support the artist somehow feels... cumbersome and backwards. Setting up a donation alternative can't be that hard, and considering Trent himself paid $5000 for the Radiohead album... Why would he conclude that noone would do the same with the (brilliant) Saul Williams album?

Mark Harrison

1: I can't understand why this is disheartening IN THE SLIGHTEST.

Conceptual model 1: The people who took the free download option are the equivalent of those who would have pirated the music beforehand. On that basis 20% of people paying, 80% pirating feels about "normal."

Conceptual model 2: The people who took the free download option are the equivalent of those who previously would have heard a few tracks on their local radio station. On that basis 20% conversion to purchase feels, well, unbelievably good.

The thing that I suspect needs price testing is whether $5 is the right price point. How about a "lowish quality" for free, $3 for medium quality, $8 for "CD-quality" [1]

[1] I'm talking like a marketer, not a technologist. You know what I mean :-)

2: In terms of free downloads, surely the upgrade path is to capture an (optional) method of communicating with the customer? In any other business, the life-time value of a channel to 154,449 self-selecting customers would be enormous. (Or 28,322 depending on how you count.)

I suspect that more than 50% of the customers making the purchase would have agreed to either receive an email "free newsletter", or subscribe to an RSS feed of the same.

The money's in the live gigs and merchandise, now, isn't it, and THAT'S about building a FANBASE... or have I fundamentally misunderstood?


Quantity or quality?

Even though there were less that paid for the album, he kept more of the money. In doing so, it also got his art in the ears of so many that probably wouldn't have heard of the guy. I downloaded the album and didn't like the album as a whole, but did like some of the songs. Plus, if I hear his name I would recognize it. Can't say the same thing would have happened a few months ago!

It is marketing 101... just a chance to get your message (music) in front of potential customers. If you are in it for the business, and didn't loose money (looks like he will have made a few dollars), then how can you lose?


Chris, you're mischaracterizing Reznor's post to make a point. He's far more optimistic than you make him out to be, and makes the valid point about Saul's increased visibility and exposure.

It's also a little distressing to see that you only quote one word from his post, and it's misquoted -- he does say that the results "seem disheartening" before qualifying that with some positives, but he never describes himself as personally "disheartened" as you quote him doing.

Reznor's post is pretty lucid and balanced, even if he does seem to be leaning to a conclusion that you don't agree with. You can do better covering this one, Chris.

Aaron Ouellette

I can't say that was a good judge of the future of the medium. I listened to the album for free, and it was not worth buying the full version, and not even worth me keeping, as I've since deleted the music.
I don't think it's an issue with the distribution but rather the content. Since we are dealing with something that you can't directly compare (album to album sales / downloads) we would need a Super Cruncher sized set of figures in order to accurately predict the viability of this medium. If only we could accurately calculate the comparison of i tunes and other online music stores, vs the free sources (given they were legitimate).


Thank god I did my download for free!
I hated all the songs....
Radiohead was another story. I paid my US$5 with a smile on my face.

Márcia Lira

Hi, Chris. I work for an advertising agency's magazine in Brazil (www.ampla.com.br). We'll write about the long tail theory in the next edition and it would be wonderful if we could make an interview by e-mail with you. What do you think? Please, send me an e-mail. Thanking you beforehand.


I downloaded the free version and was planning on happily paying the $5 if I liked it but this album is a departure from William's previous sound and it's not one I like.. I've bought a couple of his previous albums and would've been a little upset if I had forked over money for this one.. so while he might not have gotten my $5, he does have my good will because I was able to sample new sound without having to pay for it (especially since his music is never/rarely on radio)


i have to agree with ryan. trent might want to check out this article. apparently most folks can't tell the difference after 128kb.


Beyond the numbers, there is always the issue of the actual music that needs to be take into account: maybe it sucks...


personally i find the figures interesting more than anything else

i'd be slightly disappointed if only 20% of the people who sought out my work and wanted to hear it thought it was worthy of paying for.

it's also important to remember that this got a lot of free publicity as being one of the first albums distributed in this way. if it's the way of the future (as many say it is) then the market will be flooded with similar releases and a lot fewer than 150,000 people will get to hear it. and if it's still only 20% of people that think it's worthy of paying for then it doesn't bode particularly well for independent artists

Andy Baio

Also, Saul sold nearly as many copies of Niggy Tardust in the first two months as his previous album did in the last 3+ years. I'd love to see a breakout of actual artist revenue per month between albums.


He's not disheartened by the numbers. He's disheartened by the business model.

Usom Yanon

This is pretty far below the fold, so I don't know if anyone will read this. Here it goes anyway:

"[M]aybe there's a way to release stuff like this via torrent, so that the artist doesn't bear the sole cost of the bandwidth."

I'm pretty sure TR went this way so he'd have some semi-reliable data on how many people downloaded the free version. A lot of his fans would have seeded a torrent on principle, but (disclaimer: talking out of my ass) I don't know that you can really track how many people downloaded a torrent file. $0.02.


I didn't read all of these comments, but you also have to factor in that a LARGE portion of the people that downloaded the free album were NINE INCH NAILS fans. Not Saul Williams fans. That being the case, many people probably wanted to know what Trent was so interested in thinking it might be similar. When they found out it was anything but they probably deleted the album right away and forgot about it. Personally, I prefer an album in a case with artwork because burned CDs look jank. However, if Trent released a NIN album via download only I would not even bother with the sample version. I'd pay for the flac version and not think twice about it.


I understand that Reznor might not have been totally negative. Otherwise, I'd say that it's pretty disingenuous to offer your product for free and that get mad at people for taking you up on it. The cool thing about the Radiohead promotion was that there was at least the appearance that they were totally fine with you taking it for free and that they were really appreciative of people who were willing to donate or pay for it. When you offer it for free but then expect people to pay, my guess is that people can catch that vibe from you and may affect whether or not they want it.

John Lambie

Hmmmmm. the way I do the math he got:
a) double the cash
b) triple the fans, and
c) 154,449 times the intimacy

Add to that the awesome media exposure, what's to lose?

Peter Kohan

The other thing to remember here is that the results Trent and Saul produced was over a fairly small slice of time, just a few months. Saul can continue to build his fan base and earn more revenue from his recorded assets than on the previous model he was working under.

But this model isn't for every artist, and it could be tweaked in a myriad number of ways.

What Trent is disappointed in is the clear moral line he feels many have crossed in paying nothing for Saul's album. It's that 80% saw Saul's album as low-hanging fruit to be shoplifted.

Why should artists have to bear the brunt of the financial impact because consumers have decided to abandon their sense of moral clarity regarding the concept of theft and basic property rights?

Justin Boland

"Why should artists have to bear the brunt of the financial impact because consumers have decided to abandon their sense of moral clarity regarding the concept of theft and basic property rights?"

BECAUSE NOBODY ELSE WILL DO IT FOR THEM. If your business model requires laws to keep it profitable, GET A NEW MODEL.


RE: But in the direct download model, which bypasses label distribution, the artist can keep everything, as Radiohead did. I don't know how Reznor and Williams are splitting the money, but between them they made $142,000 this time, some two-and-a-half times more than Williams did last time.

This is not really true, at least not for In Rainbows. Radiohead's contract with Warner/Chappell meant that they had to pay WC for every download, even the ones given away for free. They didn't even know this until shortly before they did the Internet release, as is described in the second conversationwith David Bryne on Wired.

My understanding is that these payments for In Rainbows were about half a million Pounds Sterling and a fair fraction of the total revenue.


Recording studio costs? Trent has to have a studio at his house. He certainly must host his website at an ISP - so only incremental bandwidth charges. Nice tax write-off to offset his revenue. Plus all the blogs, interviews equals free promotion/marketing. After all, would anyone know who Saul is?

Moreover, it seems that audio fidelity has lost some of it's cache - back in the day we used to care about such things as frequency response - now that we listen to music on youTube, we think that iPod audio is great. And all of us old rock-and-rollers are so deaf now that we can't really tell anymore.


Is anyone else old enough to remember record stores with sound proofed listening cubicles? What percentage of listeners bought the record?

Some of us even remember radio. What percentage of listeners bought the album? In any form?

How many people buy music without having heard at least a track sample? How can an artist get someone to hear that first sample? If you have the bucks, you can buy radio time from the radio company. You can play it on the street or in a club, but how many people will hear you?

There's an old Japanese show business saying, "If you have three strings, you will eat." (Of course, they say it in Japanese). If you want to do more than eat, like buy replacement strings, you have to do more than play.


I bought Radiohead's album. I wouldn't have bought it but for its easy availability online. Truth is that I haven't listened to it even once. Radiohead got a couple of dollars from me that they wouldn't have otherwise got.


I bought Radiohead's album. I wouldn't have bought it but for its easy availability online. Truth is that I haven't listened to it even once. Radiohead got a couple of dollars from me that they wouldn't have otherwise got.


thank you



accessoires xbox 360

Well done analysis, but I think you could further it by providing more insight on how the free downloads could eventually be converted into real revenue. You can estimate that with Radiohead’s “In Rainbows” that it was downloaded illegally by over 5M people. If even 1% of those people become real fans and pay for a concert ticket in their lifetime they can make quite a bit of money. I think this is where the power of freemium lies.

The comments to this entry are closed.


The Long Tail by Chris Anderson

Notes and sources for the book

FREE was available in all digital forms--ebook, web book, and audiobook--for free shortly after the hardcover was published on July 7th. The ebook and web book were free for a limited time and limited to certain geographic regions as determined by each national publisher; the unabridged MP3 audiobook (get zip file here) will remain free forever, available in all regions.

Order the hardcover now!