Posts from

Taste Neighbors (continued)

A few summers ago, I penned a post called Taste Neighbors in which I described a web service that would do for food/restaurants what has done for music. From that post:

This problem has largely been solved in music. Because its relatively simple to watch what music I listen to and what music millions of others listen to, there are many services now that use musical neighbors to drive recommendations. My personal favorite of these services is and this is a list of my musical neighbors.


I am more optimistic about watching what they actually do. As my former partner Bliss used to say, "watch what they do, not what they say". With online finance services like Wesabe (one of our portfolio companies), you can easily build a database of every restaurant you eat at, every movie you go out and see, every book you buy from Amazon, etc.

Well Wesabe has come and gone but my interest in a taste neighbors service has not.

Last night my friend Vanessa said to a few of us, "couldn't you look at my foursquare checkins and figure out what other foursquare users like to go to the same places I like to go and then using their checkin history, recommend other places I might like to go?"


So is anyone doing this? Can it be done via the existing Foursquare API? I think this is a big idea and I'd like to see some people working on it.


Airport Express and Airfoil

We've had all sorts of digital music systems over the years. And they keep getting cheaper, simpler, and better.

We started over a decade ago with an over the top multi-room system that was built on Audio Request music servers and controlled by Crestron. We scrapped that system about five years ago (although I kept two of those music servers for each home we own). We moved to a Sonos based system. We still use that in the main rooms of the homes we own.

But my current favorite music system is an Airport Express connected to a simple and cheap amplifier and speakers. We use that in all of our kids' bedrooms and increasingly in other rooms in our homes like our guest rooms. Basically you bring your laptop and iTunes library and we provide the sound system. I like it very much.

But my main beef with the Airport Express is it is limited to iTunes. In the past couple weeks, two friends mentioned Airfoil to me. So yesterday I downloaded Airfoil to my laptop and gave it a whirl. Basically Airfoil intercepts any audio stream on your machine and sends it to the Airport Express.

This is the perfect freemium experience. I downloaded it and tested it by playing in my Chrome browser and it played flawlessly on the music system in our beach house. I've been wanting to have on our pool deck for years now. I've got it now. Sweet.

So of course I bought a five license bundle just now for $46. We'll have Airfoil on all of our laptops before the day is over.

Of course Apple should offer this feature built in on the Airport Express. There are a lot of things Apple should do. I've given up hoping or expecting them to do it. I'm just happy some crafty software engineers built the hack we all want. Thanks Airfoil.

When our oldest daughter came home from college this year, we had already moved into our new apartment. I showed her how the Airport Express worked in her new room. She said, "cool, how do I play radio on it." I didn't have a good answer for her. Now I do. That's progress and I'm really excited about it.

#My Music#Web/Tech

Mobile Audio

There's a reason why radio and outdoor (billboard) advertising together became a $30bn to $50bn annual domestic market. When people are mobile, like driving a car, they are not reading, they are not watching video, they are not opening email. At least they should not be doing those sorts of things while driving a car.

While radio and billboards will still be attractive advertising opportunities for some time to come, there is a new way to reach the mobile consumer – on his or her phone.

I'm not talking about calling you or text messaging you on your phone. I am talking about when you connect your android phone into your car's audio jack or when you put on your iPhone headphones and hop on the treadmill at the gym.

In these situations, you are likely listening to audio and increasingly streaming audio. That audio stream can contain commercial messaging if it is done right. And because the phone, as opposed to the car radio or the billboard, knows a lot about you, including where you are, the messaging can be targeted (ie made relevant).

This is the opportunity our portfolio company TargetSpot was built to go after. When the company was started, it decided to focus on terrestrial radio companies and help them monetize their internet streams. It is the leader in that market today. Then it added "pure play internet radio" providers like Yahoo! Music, MySpace Music, and AOL Radio to it's network and further solidified its lead.

And today, TargetSpot is rolling out its first mobile audio advertising service, in partnership with Slacker. If you want to reach people who are listening to streaming audio via their phones in their cars, in the gym, at work, and at home you now can do that via TargetSpot.

Slacker is one of several streaming audio companies focused on the mobile phone. Others include Pandora and I expect we'll see hundreds of providers over time.

And I expect that we'll find out that the audio format is one of the most powerful forms of mobile advertising. Just like it has been in the offline world for the past century.

Enhanced by Zemanta

#VC & Technology

From Scans To Listens

For years the number that musicians and their business partners, managers and record labels, would focus on was the soundscan results. From the soundscan website:

Sales data from point-of-sale cash registers is collected weekly from
over 14,000 retail, mass merchant and non-traditional (on-line stores,
venues, etc.) outlets.

My friends in the music business would always say things like "we scanned 25,000 units last week". In a world where the sales of CDs (and before that albums) was the key goal, it made sense. Scans were revenues.

But the world has changed and more and more music is available on the Internet for free or via subscription services every day. And just because you scanned a record, doesn't mean you listened to it.

I think it is time to stop focusing on scans and start focusing on listens. But how do you do that?

I bumped into my friend Spencer Hyman yesterday at breakfast. Spencer used to run I asked him if you could use as a panel, like Nielsen or comScore, to measure music listens. He said you could but you'd need to do some statistical weighting by geography and genre.

If you look at the charts for 2009, you'll see that Lady Gaga was most listened to artist on with 755,000 different listeners and 18.5mm "album scrobbles". If you go to the Lady Gaga page on you'll see that her song Poker Face has been listened to 235,000 times in the past six months.

If you think of the people who scrobble their listens to as a panel, then you can scale up these numbers to get to worldwide listens. hasn't done this work but they should. They could be the new soundscan as the key metric moves from scans to listens.

As an example of why this metric will be increasingly valuable, let's go back to that Lupe Fiasco mixed tape that includes the HP advertisement as a track in the mix. That advertisment track is called HP Skit. says HP Skit was listened to 3,850 times last week. Let's say 50x is the right multiplier for this artist and song (it's an educated guess). Then HP Skit was listened worldwide about 200,000 times last week. If that ad is worth a $20 cpm, then Lupe could have earned $4000 in ad revenue last week.

All of that is hypothetical to some degree but hopefully instructive. Music is moving from a physical good where scans is what matters to a virtual good where usage and engagement matters. So let's start measuring it correctly. That may help the artists get paid correctly.

Reblog this post [with Zemanta]
#My Music#Web/Tech


I use a lot of music web services but I don't like to invest in this sector. Nonetheless, it's an area that I spend a lot of time thinking about. I've written endlessly on this blog about the music web services I use, why I use them, and where I think the music web is going.

The most interesting music web service to me has been audioscrobbler (aka I'm not all that interested in as a social network, but I am obsessed with it's value as a data asset. I report all my music listens to via the audioscrobbler technology and it has built a deep data asset on my musical listening habits (and therefore musical taste). Since October 2005, I've recorded 60,168 song listens with audioscrobbler. That's roughly 40 listens per day. Sounds like a lot, right? Well we listen to music all the time in our house and we've had audioscrobbler on our Sonos for the past year or so.

There are a bunch of music web services I use that leverage the power of the audioscrobbler data via the api. So I show up at a new music web service and it can instantly know what I like to listen to by simply asking me for my user name and password. It's like magic. I love it.

The developer of audioscrobbler is a guy named Richard Jones (aka RJ) who built it while he was in college. He merged it into and became the CTO.

Well RJ is back to building interesting new web music stuff and his new thing is called Playdar. And like audioscrobbler, I think this could be a powerful foundational platform technology for the music web.

Playdar is a "music content resolver" platform. You put the Playdar software on all the machines you have with music on them. And then Playdar makes it so that you can play your music via the web whenever and wherever you want. This is not the first effort to do this sort of thing, but it is the first time this has been done as an open source platform.

This is an important distinction. Like audioscrobber was the foundational technology for and many other music web services, Playdar can and will be the same.

The Playdar ecosystem is just getting going but there are already some interesting demos. I like Toby Padilla's Playgrub which turns web pages into playlists. I also like James Wheare's Playlick which turns accounts into playlists.

Open platforms and ecosystems are powerful and the music web needs more of them. I am excited to see where Playdar goes. I'll be following it closely and if you are into web music, you should too.

Reblog this post [with Zemanta]
#My Music#Web/Tech