|Radio Airplay 101 - Personalized Online Radio:
Many artists have been trying to promote their music by using personalized online radio sites. These sites are basically "radio stations" that you subscribe to, and you tell them what kind of music you like, so they can play those songs and other "similar" songs. These sites say things like "we play the music you like". Now, these sites are indeed the BEST places to listen to cool stuff that you like, but they are the WORST places to PROMOTE you own music. Why?
Let's first look how these sites work. Let's say a site has 10 users (listeners), and 10 songs. The "listeners per play" ratio is 1. This is because 10 listeners divided by 10 songs = 1. So a single "play" of the song only gets you a single "listen".
Now, say that a regular AM/FM/Satellite station has 10 listeners, and plays 10 songs. When a song is played here, it get 10 "listens" right away, because all listeners hear the same song. After all 10 songs have been played, all 10 songs got 10 "listens" each. So a single "play" of the song gets you 10 listens instead of 1. (How many "hits" are made by people hearing a song only once?)
Here's the problem: On the sites, even when artists get huge "listens" (like two million), nothing else happens. In other words, even after a song gets 2,000,000 "listens", there are no CD sales, no merch sales, no contacts from labels, stores, managers, clubs, or anything else. How can this be, if the song was really listened to this many times? Here is why (some or all of the following are happening on these site, depending on the site)...
1. There really aren't that many "listens" in the first place; the number could have just been made up. A site owner can do whatever he wants, including telling you that you have ten times the number of "listens" than you really got. There is no way to know which site does it, and to which user, but it's easy to do. Just like it's easy to change the number of "plays" on a myspace page if you know how to upload code.
2. The "listens" are scattered over the whole world, and thus no particular city develops a "buzz" about it. Much less in your own home town.
3. The "listens" could really just "playlists", and not "plays". Being on a playlists is of no use unless someone actually listens to it.
4. And here is the big one: Nobody, on average, hears your song more than once:
One large site says they have over 20 million users, and over 600,000 songs. BUT!... this means that when a new user selects a song to listen to, you have a 1 out of 600,000 chance that he chooses yours! And the more songs there are, the worse it gets. This is a critical difference to understand when comparing it to AM/FM/Satellite. So, using their own numbers, you have 20,000,000 listeners divided by 600,000 songs = 33 users/song. This means that on average, each songs gets 33 "listens" by 33 different people, and these listens are scattered all over the world. Even worse, each listener is hearing it only once! And the more songs a site get, they less your song gets heard, because the listeners-per-song (# listeners divided by # songs) ratio get less. And what about the example of one song getting 2,000,000 listens? Sure it may be at the top of the stack if you are counting "listens", but there are 20,000,000 users, so that means that 2,000,000 listens divided by 20,000,000 users = 0.1 listens per user! Yes... each user, on average, hears the song only 0.1 times... one tenth of one time!
Compare this to a club: A DJ plays a song in a club with 50 people, and everyone there hears the same song one time. That's 50 "listens". But it's just one song (yours!) getting heard by 50 different people. So you have one song being heard by 50 people, one time each. The number of "listens per song" is 1, which is ten times higher than the above example of 0.1
Next, compare the sites to a record/cd store. Say there are 1000 different songs on the shelves, and your song is one of them. Also say there are 100 people that come in to shop. You have a 100 out of 1000 chance that yours will be bought, and thus your "listeners per song" ratio is 100 divided by 1000 = 0.1
Now say that they double the number of songs in the store (Wow! Great!) to 2000. Well, now your chances are cut in half: You now have only a 1 out of 2000 chance that your song is chosen. This is why the more songs there are, the worse it gets for you.
Compare all this to AM/FM or Satellite radio: One large New York City radio station has 2 million listener (12+ cume 6a-12a mon-sun arbs) and plays 40 songs. This means (roughly) that each "play" of a song gets heard by 2,000,000 people. Thus, if a song is played one time, it has 2,000,000 listeners, or "listens". If that same song gets played 20 times a week, it has 20 X 2,000,000 = 40,000,000 "listens" in that week. And this is exactly want you want: One song (yours!) heard by lots of people, lots of times. (Especially in just one city, where everyone will talk to each other about it.) And the important number you get from all this is "listeners per song": 2,000,000 listeners divided by 40 songs = 50,000 listeners per song, instead of 0.1
You say, "Fine, but there are so many people on the sites, that tons of them have to hear my song!" Nope, won't happen. The only way to make this happen is to cut the number of songs available down to almost nothing. Even if yours was the song that got 2,000,000 "listens", you would have to reduce the number of total songs from 600,000 down to 40 (like the radio station example), in order to have a listener-to-song ratio of 50,000. And this is not going to happen.
Click here for a list of all topics...
Next topic: How To Use College Radio
For a complete description of our airplay campaigns, including pricing, send an email to email@example.com