Matthew Ogle and Anil Bawa Cavia – Lessons from Building the world’s largest social music platform

Last.fm is very popular with the younger members of our house, although some of the older residents are perhaps getting a little tired of hearing S Club Seven blaring out of the computer in the living room! It works nicely on laptops in hotel rooms, too, bringing a little bit of the real world to the dreary corporateness of a Hilton or a Holiday Inn…

Matthew Ogle is up on stage here at Future of Web Apps, talking about some of the lessons they learned in building and growing this UK-based social media success story.

Some numbers;

15 million tracks scrobbled per day

175 scrobbles per second

Over 6 billion tracks scrobbled since 2003

10 million artists

70 million tracks

700,000 30 second clips streamable via the Last.fm radio

17 million items tagged

145,000 wiki pages about artists

Early growth lessons;

• Don’t overextend – scale with growth, not before
• Make sure the revenue model scales as usage increases
• Involve users in you and your story; make your growth a selfish aim for your users
• Be as open as you can afford to be

Openness and growth

Early audioscrobbler protocol, to consume data from established audio players, rather than trying to persuade users to switch to a new and content-light service. Incentivise developers, and allow users to gain value with little pain.

Promote a community around your application, and talk to your users even when the news is bad.

Anil is talking about the realities of growing a small company into a larger one, identifying the importance of appropriate processes as developer teams grow. He also makes the point about the importance of opening up apis and other means of access, to speed the move from ‘application’ to ‘platform. Familiar!

Back to Matthew, who’s talking about the difference between scrobbling (’myware’) and other tracking technologies which tend toward the more traditional ’spyware’. He’s making reference to the Attention Economy (see an early podcast of ours with Ed Batista of the AttentionTrust), and pushing the value of leveraging your own behaviour (what I listen to), combined with the aggregate attention flows to create something of value to the individual and the community. [sorry - hard to string a coherent sentence together whilst listening...!]

Anil is talking about ways of making money on the back of the Last.fm model; Monetising Attention in essence. He’s referring to a blog post by Fred Wilson, which he summarises thus;

microchunk it – reduce content to simplest form

free it – put it out there without walls around it or strings attached

syndicate it – let anyone take it and run with it

monetise it – put the monetisiation and tracking systems into the microchunk

Some interesting stuff about leveraging attention data to clean up tag clouds, etc.

Last.fm plans;

More – growth, streamable music, ambient findability, personalisation/ things you can do with your data

Less – interfaces, barriers to entry, gradients

And yes, Last.fm will be an important part of tonight’s hotel room stay… if there’s affordable wifi!

Technorati Tags: FOWALondon07, FOWALondon2007, Last.fm, Social Networking, Talis