Skip to main content



Diffusions in the Music Streaming Industry

https://www.fool.com/investing/2018/10/16/googles-plan-to-compete-with-spotify.aspx

This article details Google’s plan to compete with Spotify in the music streaming industry. As many people are well aware, Spotify has been the dominant player in this industry for some time. Naturally, because all streaming services provide pretty much the same product for relatively similar prices, the streaming business proves to be a difficult one to break into. However, Google has something going for it that Spotify, Apple, and all other competitors do not: video. Google owns YouTube, and Google plans on using YouTube for its music streaming service. For free, all users of YouTube Music will be able to download and view music videos on their computers or mobile devices. As I mentioned before, because the music itself and its price is similar to Spotify, this feature should help to give Google a leg up on its competition. It’s simple economics: if a better product is offered for the same price, consumers will respond accordingly.

But, in this case it’s pretty clear to anyone who follows the music industry that YouTube Music will do little to loosen Spotify’s grip on the music streaming industry. To explain why, it helps to bring in a comparison to network diffusion.

Imagine a network of music listeners. This could be posters on a forum like Reddit or readers of a music publication like Complex. Every node in this network has a choice between using YouTube Music or Spotify to listen to their music. Right now, I think it’s definitely fair to assume that the whole network uses Spotify. As a result, the forum/publication will use Spotify links in their discussion of music. If I were to join this network as a YouTube Music user, I would constantly have to look up the music being discussed. This hurts my utility as a YouTube Music user. It doesn’t matter what kind of perks YouTube Music offers; it makes much more sense for me to join Spotify in order to maximize utility in consuming music news.

This raises the question of whether diffusion is even possible in this network where publications and consumers lean on each other in their Spotify usage. We can take a look at the three ways to spread diffusion to see if any of them could realistically happen in this scenario. The first suggestion is to increase the number of nodes/links in the network. In this scenario, everyone is more or less linked to everyone else, so adding more links is unlikely to work if it is even possible. However, if enough YT Music nodes join the network and place pressure on publications to add YT Music links, we could see diffusion happen. This argument also works with the second suggestion of increasing the number of adopters. The third suggestion of lowering the threshold q is difficult to put into practice in this situation. There is no real way to affect consumer preferences if you’re Google, even with ads. However, a couple well-placed dollars could help influence forums and publications to change what service they link to. This could work, but at the end of the day consumers could always just stop visiting the site in favor of visiting one that uses Spotify. It would be high-risk for publications to unilaterally switch. Basically, it’s easy to say that Google faces a costly uphill battle in swaying consumer preferences of music streaming services.

Comments

Leave a Reply

Blogging Calendar

November 2018
M T W T F S S
« Oct   Dec »
 1234
567891011
12131415161718
19202122232425
2627282930  

Archives