Hi guys, welcome back to another week of Web 2.0 thrills and spills. This week we’re moving onto the 2nd pattern described by Mr O’Reilly – Data is the next “Intel Inside”, against which we will analyse Spotify.
For those of you unfamiliar with Spotify, it’s an audio streaming service that allows you to search and listen to just about any band, musician, composer, orchestra or comedian entirely for free. You can pay for an ad-free, higher bitrate and offline service for a nominal fee, but if you don’t mind having the odd advert play now and then and you’re cheap (like me) then the free service is awesome. Spotify integrates with your Facebook account, which lets you instantly share songs with friends and posts your activity on your time line.
So let’s kick off the analysis by first understanding what this week’s “pattern” really means. Simply, O’Reilly is stating that the key ingredient in web 2.0 applications is the data that is presented to the user and how this is done; whether that is a YouTube video, a seller rating on eBay or a song on Spotify. These are known as data strategies or business models, and there’s a variety of different ones.
Creation – where data is created by the users themselves such as on Facebook
Control – where access to the data is strictly controlled through various permissions or file types much like how Spotify lets you listen to any music you want but only through specific account types are you able to access the data offline and currently there is no way to “burn” or “rip” a song onto disc.
Framework – such as what is provided by Google’s location services, email and search facilities
Access – which can be defined as utilising data that is created by a third party and displayed in a new or novel manner, for example the Runkeeper iPhone application, utilises Google maps to record, analyse and display distances and routes that you’ve run.
Data Infrastructure – is where an infrastructure is provided for others to store and access data, such as the Windows Skydrive.
As we’re looking at Spotify I want analyse their business model against some of the best practices involved in this week’s pattern.
Seek to own a unique, hard to recreate source of data
Apart from artists such as The Beatles and Led Zeppelin, Spotify has just about every single song you could ever want to listen to. Whilst these two bands are arguably the biggest on the planet, to only have them missing from what is seemingly an endless source of music and audio entertainment, certainly makes Spotify’s data source not only unique but insurmountably hard to recreate.
Enhance the core data
Spotify’s radio functionality which works on a proprietary algorithm that effectively creates a mix-tape of songs based on a song’s genre, sub-genre and the “thumbs up” or “thumbs down” inputs of users (much like stumbleupon’s stumble functionality), ultimately enhances the core data by improving the algorithm based on the types of songs that the user wants to listen to. Furthermore users can favourite (star) each song which ultimately improves a song’s popularity which benefits the artist as well as the user.
Let users control their own data
Spotify has a variety of functions that lets the user control their data, for example if you don’t want to post information about your show tunes playlist on Facebook you can switch this off. In the premium version you can make all your favourite songs available offline, in case you’re ever in an area that isn’t serviced by internet or 3G for example.
Make some rights reserved, but not all
Copyright is a pretty hot topic in the music business and has been for years and you could be forgiven for thinking that Spotify’s basic business model would wipe out music piracy and solve all the issues facing the music industry, but whilst it is a giant leap forward it hasn’t entirely ensured that it actually benefit all musicians.
Design data for reuse
Navigating the Spotify user interface is often frustrating and unintuitive, and whilst their search function often delivers you the song or artist that you’re looking for. Knowing which button to press or what is a hyperlink and what is not, however, is at times infuriating. With such an overwhelming amount of data available this is one element that they really need to work on. It seems as if they’re trying for an iTunes look and feel but I’m not sure that it is an effective analogy for what Spotify is, which is more than just an audio streaming service but an online database of just about every artist on the planet.
Music piracy is the fundamental issue that Spotify’s business model have resolved but whilst there are still issues to do with how all artists featured can benefit fairly. I firmly believe that we’ll see more of these types of subscription based models popping up for more and more media based products. Netflix for movie downloads is already one such service, but I would expect to see live sporting streams to become part of this same business model as well as TV shows. Both of which are affected negatively by pirate streams and P2P sharing.