iTunes Song Purchases plotted on a graph

Apple recently announced 1 billion songs purchased on iTunes. I dug around the Apple press releases and tracked down some of the major iTunes milestones over the last few years. It’s amazing that there have been 500 million songs purchased since July 2005. That’s 500 million songs purchased in 7 months for an average of 71 million songs purchased per month!

2/26/06 Update:

Additional data point: 850 million on Jan 10, 2006 (announced at Macworld)


Songs Purchased (millions)


05/15/2003 1 Apple
06/23/2003 5 Apple
09/8/2003 10 Apple
12/15/2003 25 Apple
03/15/2004 50 Apple
07/12/2004 100 CNN
12/16/2004 200 Apple
07/17/2005 500 Apple
02/24/2006 1000 Apple

graph created using NCES Create a Graph

Filed under: ,



Filed under Uncategorized

44 responses to “iTunes Song Purchases plotted on a graph

  1. Anonymous

    Those are incredible numbers. I wonder if they are real?

  2. Anonymous

    Of course its real, apple just reported the 1 billionth song download.

  3. anonymous #1 – this graph is based on official numbers from Apple or reliable new sources (see links under Source column). As a public company they can’t get away with exaggerating these numbers.

  4. Anonymous

    Wow, I would never have I would so such a graph, thanks for the effort, now I suggest you get back to what you were doing before hand.

  5. well, they can, it’d just be illegal. there is no real adavntage to faking the numbers at this point anyway. at the beggining it may have made sense to make people comfortable, but by now, people recognize iTunes as a legitimate source for purchasing music.

    it’s interesting, and suprising, that they hit 1 million so fast (around 1 week).

  6. When did iTunes open?

    Your plot starts with 1million d/ls I wonder how long it took to get even that.

    (The longer the time for the first million the more impressive the recent growth is… in a way.)

  7. Anonymous

    …unless you’re ENRON.

  8. Anonymous

    I’d like to see this graph overlaid with iPod sales figures.

  9. Paul Graf

    You forgot the 850M from MWSF 06

  10. Po

    850M Watch the MWSF Keynote

  11. I’d like to see this graph overlaid with iPod sales figures.

    I would be willing to bet all I’ll make this year it’d they’d be strikingly similar.

  12. Anonymous

    I’d be interesting in seeing what the top 50 downloads were?

  13. Thanks Paul – I have added a data point for the 850 million announced last month at Macworld.

  14. Ookami – they hit 1 million downloads after 1 week of operation.

  15. Thats mighty impressive, it will be interesting to see if music sales keeps this up! I wonder how the video sales will compare… way to go apple!

  16. No one commenting on the exponential curve of the graph? and you thought you would neve use math after high school.

  17. For those of you questioning the numbers, kudos to you. Apparently some people have already forgotten how definitions are like statistics and can be manipulated. Case in point, in the 90’s Microsoft redined the term ‘shipped’ to mean the number of units that had left the manufacturing plant (to the warehouse). Previously the term meant the number shipped to the customer. I’m not sure how the term ‘purchased’ can be twisted. How about other folks?

  18. Regarding Tom’s observation, does anyones know if Apple is counting video purchases as “songs”? Which reminds me, iTunes is kind of a dumb name for a marketplace that sells music AND VIDEO.

  19. Anonymous

    It would be really interesting to know the number of songs PURCHASED vs. DOWNLOADED. i’m sure apple gives away a lot of coupons… especially when they are trying to hit milestones before events… coincidence that they hit the 1 billion downloaded number a few days before their 2/28 event? no way.

  20. Anonymous

    Um, we don’t have to jump immediately to exponential do we? It is certainly nonlinear but it could easily be quadratic or something else less extreme. It does approximately double over a 15 week period for weeks 30 to 45 and 45 to 60. After that the doubling period is longer. If it were exponential there would be a constant doubling period. Just my own minor attempt to prevent exponential from being overworked.

  21. Anonymous

    Perhaps Apple plotted ITMS sales and scheduled their event about a week after the billionth song would be sold. I think they do have a couple people in the company that can do arithmetic.

  22. Anonymous

    I suspect that Apple counts the free song download of the week as part of their sales figures. The reason? During one of the early sales milestone contests (maybe 100M), they published a list of the winners and what song they had download. Among them was a guy that won and the song listed was the free song for that week. This was pointed out on and then they covered it up (either removed the name or the song title) by the next day. So it would be very interesting to see how many of the 1B songs were the free songs of the week, because I’ve downloaded just about every one of them for over a year but have only purchased about 15 songs.

  23. If this numbers are true, why the hell is the RIAA complaining…

    It’s bullshit.

  24. Extra Character

    The title of the graph says Purchased Songs. Which, I assume excludes the freebies. Even if their figures are off by, say, 5 million I think it is still an impressive feat.

  25. Regarding some comments about the validity of the numbers, take a look at Apple’s 10-Q report. On page 26, you’ll see $ 491 million for Q4 2005 (it was $ 177 in Q4 2004) for “Other music related products and services”. I think those numbers back up what is shown on this graph.

  26. If you read the sources, it’s the number of songs DOWNLOADED, not the number of songs purchased. So the title of the graph is inaccurate, but other than that…

  27. Astonishing that so many people are prepared to buy expensive, low quality tracks that are infected with DRM and only play on one vendor’s platform.

    Which raises the question that a much more intersesting stat would be the growth in customers and the changes in tracks per customer.

    Just Say No To DRM

  28. Anonymous

    What would be even more interesting would be some addition marking when each additional store opened. Each store will have a growth curve of its own that adds up to this. Would be nice to know a store by store breakdown too, but I can’t see apple releasing that kinda info.

  29. To put this in perspective, at Macworld last month, Steve Jobs announced that they are selling 3 million songs per day, worldwide — a run rate of more than 1 billion songs per year. Compare that to week 1 when they sold 1 million songs in 1 week.

  30. Anonymous

    “Astonishing that so many people are prepared to buy expensive, low quality tracks that are infected with DRM and only play on one vendor’s platform.”

    No what’s astonishing is that you geeks still think the general public cares about any of that.

  31. Anonymous

    “No what’s astonishing is that you geeks still think the general public cares about any of that.”

    The general public *will* care when the discover the deal they’ve made with the devil.

  32. But the devil is so pretty! He’s made it so sleek… and it’s white, how can it be evil?

  33. Anonymous

    i wouldn’t consider iTunes the devil. I’d consider Napster the devil. Once you buy a song on iTunes it’s yours forever. Buy a song on napster and once you stop paying your monthly fee, it goes poof! Now thats evil.

  34. Anonymous

    THis is funny because it so closely resembles the ‘ideal’ sales chart that you see in cartoons in the background in some office scene or something.

    Life imitating art eh?


  35. Anonymous

    How to confirm these numbers:

    – people will be downloading ‘hits’ give or take

    – get number of ‘hits’ during period in question

    – using stats on audiences (radio stations, mtv, etc industry trackers have these) get # of fans per ‘hit’

    – using a % of fans that download, get a rough estimate of # downloads per ‘hit’

    – multiply above by # hits and voila you have confirmation ..

    – of course, you could always just believe itune’s claim : )

    – or you could divide stated gross revenue for downloads by $cost per song

    – the point is .. QUESTION the man!

    – 9/11 also requires verification (the ‘official’ story is a lie)

  36. Several people asked about exponential growth.

    Interestingly, if you plot the sales data on a logarithmic y-axis, you see that there’s a huge surge right when it starts, then a slightly slower exponent for ~2004, and now a slightly slower exponent (but still exponential) for ’05-’06 (so far).

    I think that’s pretty unprecedented — some form of exponential growth over almost 3 years!

  37. Anonymous

    unlikely to be a true exponential; more likely the first part of a logistic.

  38. It would be interesting to see the same sort of graph for iTunes video sales. I have wondered for some time about music videos. Are they ads, or artistic works in their own right? If you consider them to be advertisements for the songs they highlight, is it not ironic that people are paying to download and watch them?

    And what of video content (e.g., TV shows) which contains ads? If users are paying for their video content, should it contain any ads?

  39. An obvious reason for the shape of the curve is the fact that over time Apple has opened iTunes stores for multiple territories. From the 1 billion web site:

    “Music lovers like you in 21 countries around the globe have purchased one billion songs from the iTunes Music Store.”

    A graph charting just U.S. sales would show a curve that’s flatter than the worldwide sales curve.

  40. If anyone knows if there are any published reports showing a breakdown of of U.S versus international sales on iTunes, let me know.

  41. Tomer

    It seems like they’re in line to double sales by Q1 ’06

  42. Someone commented up above that they would like to see the curves contrasted with iPods sold.

    I whipped together a fairly crude graph of iPods sold:

  43. Anonymous

    [[Which reminds me, iTunes is kind of a dumb name for a marketplace that sells music AND VIDEO.]]

    The world is full of this — can you buy walls at Walmart? 🙂

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s