Revisiting The Long Tail

It has been nearly 10 years since Chris Anderson’s Wired article and subsequent book called The Long Tail.  In the article, book, and subsequent blog posts Chris (and I can call him Chris because we once had a 3 minute conversation in Bilbao Spain, so we are friends) showed data about how music listening is changing as we move away from the physical constraints of CD shelves and replace them with the infinite virtual shelves of the online music store.

LongTail.gif (640×418) 2013-11-23 09-10-04

I thought it was time to look at the data again to see if the trends that my good friend Chris was seeing back in 2004 still persist today. In particular, in the blog post Latest Rhapsody data and more Chris showed how a substantial fraction of the music market is shifting away from the Walmart inventory of the top 50,000 tracks:

The Long Tail - Wired Blogs 2013-11-23 08-17-04

This chart shows that as Rhapsody’s collection size increased the amount of listener market share in the songs that were not in the top 50K grew from 26%, to 28% and then 30% over 3 years.  Today’s music subscription services boast 10 million or more songs (but of course, those numbers start to get a bit meaningless beyond a certain point – it becomes filler). Let’s take a look to see if the fraction of listening that is not in the top 50K tracks has continued to grow. Here’s some pie:

Untitled 2013-11-23 08-25-24

This data shows that in 2013, with 10million+ tracks available, 42% of listens can be found in the long tail (i.e. beyond the top 50K tracks). So the trend has continued. More listening is taking place in less popular music.

In the same blog post, my pal Chris presents this Hitland vs. nicheland chart that shows what percentage of the music business is selling the top 100 artists.  Back in 2006, 50% of Walmart’s music business was selling music by the top 100 artists, while for Rhapsody, about 25% of market share was in the top 100 artists.

The Long Tail - Wired Blogs 2013-11-23 08-33-01

Lets’s extend this chart using listening data from 2013:

Untitled 2013-11-23 08-40-54

As you can see, the trend continues, only 20% of listening is in the top 100 artists.  It’s not a dramatic change, but it does show that the more music you make available, the deeper the catalog, the deeper the listening.

Here are a few more fun facts about today’s listening.  80% of all listening is concentrated in the top 5,000 artists. The top 1,000 songs account for about 13% of all listening, and 80% of all listening is spread over about 222,0000 songs.

My good buddy Chris’s Long Tail hypothesis has come under considerable scrutiny in the last few years, but by looking at the data we see that the trends Chris pointed out have continued. Our listening is less concentrated in the hits than ever before. Yes, hits are important and will alway be, but if you make more music available to listeners, they will indeed listen to it.  So, remember my best friend’s three rules for the Long Tail:

  1. Make everything available
  2. Cut the price in half – now lower it
  3. Help me find it.

At The Echo Nest, we work hard to make rule 3 a reality – our mission is to help connect people with the best music, whether it be in the short head, or deep into that Long Tail.

, ,

  1. #1 by blu28 on November 23, 2013 - 5:19 pm

    Did you see this: http://www.hypebot.com/hypebot/2013/10/the-man-vs-the-myth-lefsetz-and-the-long-tail.html

    BTW, in a coincidence that makes my head hurt, I read your post with my RSS reader, Feedly. I used Google to search for the article I remembered reading about the long tail today. After re-reading it to make sure it was the right one, I copied it and went back to Feedly. The very next unread article (which was posted after I did the Google search) was by Bob Lefsetz! I don’t even subscribe to his blog, I subscribe to a blog he was guest posting to. Too Weird.