Future Music Forum

The Future Music Forum in Barcelona this fall looks like a great European event for digital music entrepreneurs, investors, developers and music technology professionals.

Spanning over two-days, the conference will consist of panels, interviews, keynotes and workshops focused on trends, innovations and how music will be shared and monetized.

Some of the topics include: Internet radio, mobile apps and copyrights. Here's the full agenda.

I'll be attending, and moderating the panel on social media as well - looking forward to it!

Live Blogging from UGA's Music Business Program

I'm participating on a panel discussion today hosted by the University of Georgia's Music Business program, and moderated by Bruce Burch. The event is open to the public and will be held at Eddie's Attic in Atlanta; and if possible, I'm going to blog live highlights, so you may see frequent updates to this post beginning in about 5 hours. In the meantime, you can read more about the event here.

update: unfortunately the venue did not provide wireless access so we were unable to do this; hopefully next year

How Does Social Influence Effect Music Discovery?

Yesterday Bob Lefsetz wrote about the challenges of getting people to pay attention to your music. While he's correct that music recommendation and discovery is difficult, partly due to the increased amount of music content (i.e. signal-to-noise), democratically filtering out which songs and artists are the rising ones based on quantifiable and qualifiable information will require technology. For example, The Hype Machine does this well by mining the blogosphere and Twitter. The question is, will a human element remain necessary to the process? If so, what does that look like? Who are the social gatekeepers? Does their influence on an artist's popularity help or hinder music discovery? How are the social gatekeepers determined?

Seemingly, it's too early to see how social influence will impact music discovery, but I'd like to hear what artists, managers, labels and the like think by contributing to this post with their thoughts.

Free Music Business Panel Discussion April 24 in Atlanta

Save the date for a free music business panel discussion moderated by Bruce Burch of the University of Georgia's Music Business program. The event is open to the public and will be held at Eddie's Attic in Atlanta on Saturday, April 24 from 11am to 3pm. I'll provide more details soon; in the meantime, below are the confirmed panelists:

• James Andrews: managing partner/co-founder of Everywhere Social Media
• Aaron Arnold: CEO of MusicIsMyBusiness, former assistant to Sean “Diddy” Combs
• Mark "Dill” Driscoll: founder of World Sports Promotions, McCann Event Marketing, Momentum Worldwide, Ignition, and Entrepreneur in Residence at UGA’s Terry College of Business
• Chris Esposito: President and CEO of Digital College Network
• Duncan Freeman, founder of Band Metrics
• Tammy Hurt, founder of Outsource Events, Inc, and Placement Music LLC
• Clarkson Logan, co-founder of Maestro
• Richard Nailing, co-founder and CEO of Free All Music
• Doria Roberts, singer/songwriter

Also, I'll be live blogging from the event, and we'll try to provide a live video stream as well.

Band Metrics Releases Public Beta

Its been a long time coming, but I'm happy to announce that we finally released the public beta of Band Metrics this morning from Midem.

What took so long? Great question. We had to re-engineer the entire system, as the previous two versions were flawed. This was a difficult decision that consumed our time, but it's better to get the architecture right before releasing it to the public. We now feel confident we've built a system, which is entirely Ruby/Rails that will not only collect, store and analyze artist data accurately, but will be fast and scale rapidly, as we also moved our entire system into the cloud via Engine Yard. As an example, Band Metrics now pulls-in and displays initial artist data within seconds, not days or hours. Needless to say, I'm proud of our developers.

We also streamlined the site, and added a couple of new features, including Fan Segmentation and Influence Measurement. Essentially, this allows artists, managers, labels and the like to see a breakdown of fans (based initially on Twitter comments), and a fan's influence among their friends (below is a brief screencast).

We still have lots of work to do, including design enhancements and new features, as well as incorporate our existing analytical tools into the new site (we ran out of time as we wanted to release at Midem), so we greatly appreciate your ongoing support, as this is just the beginning!

Visualize the Artists you Listen to Most on Last.fm

Normalisr has made several enhancements over the last year-and-a-half, including a new site design, graphical widgets and thumbnail views of artist charts - nice work! If you haven't tried Normalisr before, it's a quick and fun app that ranks the artists you listen to on Last.fm, from the most number of tracks and time per artist, to the least.

For me, I'm not surprised by the artists Normalisr visualized for my profile, but I wouldn't rank those artists in that order. Meaning, I'd rank my favorite artists differently, but that's the beauty of apps like Normalisr, it's dynamic, like our music behavior, so it captures what we are listening to at moments in time, periods which are influenced by both internal and external variables (e.g. sociological, cultural, psychological, etc), reasons why music charts like Billboard are antiquated. Capturing and reflecting these types of influences relative to one's music listening activity has fascinated me for years, as it explores causation, an area that hopefully Band Metrics will eventually tap into.

Band Metrics Selected as one of 15 Music Tech Startups to Showcase at MIDEM

It's an honor, and I'm humbled as the competition was strong, so my sincere thanks to Midem, the judges and Music Ally for their support! Go here to read the official news release.

If you're going to Midem and would like to meetup, please send me an email.

Band Metrics on NPR

If you were listening to NPR this weekend, you may have caught a fantastic segment of On the Media called "Charting the Charts" by Mark Phillips, airing on over 100 U.S. radio stations yesterday and Saturday. It's a must listen for anyone interested in music charts like Billboard's Hot 100, as Mark eloquently explores how the art of ranking artists has evolved, and where the evolution is headed from industry leaders like Robert Levine of Billboard, Chicago Tribune music critic and author Greg Kot, Chris Molanphy of Idolator, Eric Garland and myself. If you missed the show, you can listen to it here or below if you have Flash enabled, and many thanks to Mark for his excellent research and production - it was an honor to be a part of it.

Inference is the Difference: Gaining Insight from Music Data Analytics

One of the goals of Band Metrics is to help artists and music industry professionals gain insight from the quantitative and qualitative data we collect, as identifying correlations can be the difference between increasing revenue for your band, or spinning your wheels guessing as to how your fans interact, share and listen to your music. Below are two recent screencasts that help demonstrate a couple of the areas we're working on to make meaning out of your music data (view more of our screencasts here).

The first screencast demonstrates how radio play data (soon social data) could help you build relationships with broadcast stations that are, or are not playing your music. The second one provides a suggested tour map based on the data we're collecting about your band. These are both experimental mapping features at this point, one of the reasons why we're still in private beta, but we thought you might like to see them if you don't have an account:

Real-Time Search Engines for Artists: A Quick Look at Some of the Leaders

There's been a lot of activity around real-time search during the past year, and rightfully so, as apps like Summize have readily shown the power in real-time information - it's part of what we do at Band Metrics for musicians and bands as well, as there is significant value for artists to see what's being said about them and their music at any given moment. But, there are now dozens of real-time search engines, and there are differences between them so here's a quick breakdown between today's leading, real-time search engines (listed alphabetically):

• Collecta: This may be today's present winner from the list below, as not only are the results the most recent, relevant and comprehensive, but Collecta has some nice filtering options, a solid API, and their pulling from comments in the blogosphere as well.
• OneRiot: A bit cluttered, and most of the results are seemingly coming from news topics that are shared on Twitter, but they do have an API, and I like the ability to see how a topic/story is shared by expanding it.
• Scoopler: Possibly the second best from this list, as they've got a great UI that easily separates popular content from videos, images and links, but they do not have an API yet.
• Summize: Arguably the first, and very fast, simple and accurate, but it only searches Twitter, possibly because they were acquired by Twitter - it's still my first choice for running a simple Twitter search.
• Tweetmeme: Nicely displays both tweets and online content that is being tweeted, and I like the breakdown of information between Best Match, Highest Tweets and Age, but it's a bit too Twitter-centric.
• Topsy: Possibly the least accurate of this list, as the results are not comprehensive enough, but they have a nicely designed UI, and I like the ability to easily search by day, week, month and all-time.
• WhosTalkin?: Probably the most comprehensive from the above list, as they gather information from about two dozen sources or more, but they only have a URL API at this time, and you have to refresh the page to see the most recent information, and even then it may not be displayed.

Thoughts? What did I miss?

Band Metrics Receives an Investment from the Georgia Tech Edison Fund

Happy to announce this morning that we've received an investment from the Georgia Tech Edison Fund. We're quite excited as this is a tough economy for raising capital, and we've got a ton of work to do like improving our visualizations, increasing the accuracy of our data collection, enhancing our UI, adding more data sources and features, moving into the cloud, releasing our public beta, and much more - all before the end of this year, not to mention 2010!

So, if things seem a little quiet at Band Metrics during the next 8 weeks it's because we're buried in code and building out our next release... stay tuned, as I think you'll like what we've got in the works.

You can read the press announcement here.

Band Metrics Turns One

I can't believe it, but one year ago today we announced our service from TechCrunch50 in San Francisco. While we've been working on Band Metrics for much longer, today marks our one year anniversary since this is when we came out of stealth mode and debuted our plans publicly.

Yep, it has taken us much longer to build our initial application than we had planned, but without elaborating as to why, let me just say that it's behind us, and we're now developing steadily with some exciting and powerful features in the works, and my commitment is that we'll continue to innovate and build compelling analytics for artists and the music industry. In appreciation of everyone's support, if you don't already have a private beta account, please register and I'll provide access for the next 25 signups.

Cheers!

Duncan

Characteristics of Sound Data Analysis for Artists and the Music Industry

At the core of the analytics engine Band Metrics is developing for the music industry is a Decision Support System for artists and music industry professionals, something we've been professing since the formation of our business plan. But, there are several characteristics our data analysis should have before it can draw significant conclusions about artists and their songs - this is one of the reasons why we're only visualizing artist data and insights at this time. Here are a few aims and characteristics for our data analysis, relative to the analytics we're providing and developing:

• Longitudinal: simply put, we make sound correlations when the artist data we analyze has been collected over an extended period of time, so we store all data for each artist ad infinitum
• Quantitative: diverse structured data that can be measured in terms of units, for example, the number of radio plays or fan interactions an artist has for any given time period - we're doing this now from almost ten different sources, but we'll continue to add more
• Qualitative: diverse unstructured data that aims to gather an understanding of why attitudes and behaviors are formed towards an artist, for example, why are music enthusiasts listening to a particular artist, and what are the social constructs that have influenced the fan base - we're doing the initial work for this now by pulling artist comments and reviews from two different sources (soon to be five), but we have a great deal of development that still needs to be done
• Spatial: diverse geographic data to draw correlations and patterns - we're doing this now with the Artist Mapping Platform we invented, but much more work still needs to be done
• Cross-sectional: meaning, deep and diverse artist data from within various sociographic populations and genres
• Iterative: our collection processes, data sources, statistical calculations, descriptive insights, predictive/forecasting models, recommendations, visualizations, etc. have to be continually evaluated and improved - it's a process, not an end

There are other characteristics as well. Not to mention, our data analysis needs to account for missing artist data, duplicate data, and outlier data, among other data variables. So, we're continually improving Band Metrics, as data collection and analysis is both an art and a science that is evolutionary, so be skeptical of sweeping claims about artist correlations from general observations, as the cause might be linked to something else.

Future of Music Coalition Policy Summit

If you don't know about the Future of Music Coalition Policy Summit, you need to check it out and make plans to attend this October. Speakers include:

• Senator Al Franken
• Jim Griffin
• Ariel Hyatt
• Ian Rogers
• Brian Zisk
• Jed Carlson
• Daniel Ek
• Yancy Strickler 

Topics include:

• The Post-Digital Revolution
• A Musician's Guide to the Policy Landscape 
• The License Clearance Process & Artist Compensation
• Future of the Digital Millennium Copyright Act

To view the entire list of panels, speakers and the schedule, go here. We'll be presenting as well, so hope to see you there!

Moving into the Cloud

We're about to begin our initial migration into the cloud from our existing dedicated environment so we can scale rapidly and support the large amounts of data will store as more and more artists set-up their accounts. After reviewing several solutions, we've decided to deploy Amazon's Web Services, as they provide the tools and services we need for efficient growth.

Our migration will be completed in multiple phases over the next couple of months. During the first phase, which will be operational in about 3 weeks, we'll integrate SimpleDB, improving the way we collect and normalize data, so expect more accurate data for your artists. Once completed, we'll release some of the new features we've been working on as well. Additional phases include EC2 integration and MapReduce, among others, but each phase will be spaced apart so we can continue to release new features - and we've got some exciting new releases in the works.

If time allows, I'll provide a post of what we gained by moving our infrastructure into the cloud and making these improvements to our codebase.

Two Quick Band Metrics Demos

Below are a couple of quick screencasts on using Band Metrics. The first one demonstrates how to view historical data for an artist. The second one shows how to add and manage multiple bands within one account. As we make enhancements to Band Metrics, we'll provide new screencasts. If you have any questions, let us know here.

Music and Society: Segmenting your Fan Community

Yesterday I gave this presentation at the C/O POP Music Festival and Conference in Cologne, Germany. Essentially, I explored the relationship between music and society, and how artists can better understand their own fan community through segmentation, which can lead to fan growth and increased revenue for them. Since I had several people ask for a copy so they could see the Facebook example and the resources again, here it is. Btw, Band Metrics will begin providing fan segmentation next week.

Tell Us What We Need To Know

We're now releasing new features and/or enhancements at least one per week, and we'll continue to do so from now on, but we want to make sure we're building the features you need, so please let us know what you would like to see us build. You can post your comments here or on help.bandmetrics.com.

Yeah, we know there are a couple of companies that have recently launched trying to imitate us and they've been actively following our news (even using our service), so your feedback will give them more free "product research," but hey, what can we do, as we're believers in working and collaborating with our users and the music community in an open conversation.

Band Metrics Adds Multiple Band Management

If you haven't yet noticed, we quietly released our latest product feature earlier this week that allows you to manage multiple bands through one login - it's a fantastic tool for artist management companies, labels, venues, A&R reps, promoters, music supervisors, publicists, marketing agencies and others. We're still tweaking it, so please let us know how we can improve.

Btw, this is a premium service, so please call us to get access at +1/404.492-6511.

Historical Reporting Now Live

You asked for em, so we released em - historical charts for artists - just login to your account to get started. It's only our first take so expect significant improvements and functionality over the coming weeks, but if you have any suggestions/comments before then, please let us know here. Below are a few examples of what you can now do with the historical data Band Metrics is collecting for your band (individual songs coming soon).

This example displays an overview of how an artist is trending on MySpace during a 30-day period. You can set your own time range, and/or review many other sources including YouTube, Bebo, SoundCloud and more.

You can also drill down individual sources to easily see spikes in your band's growth like in this chart:

This example displays a comparison between sources relative to your total social plays (e.g. how many people are playing your music on iLike or Last.fm. You can also set your own time range, compare other variables like fans, comments, views, and/or compare multiple sources at the same time.

With our historical trends, you'll be able to gain insights into fan activities related to your music, for example, traffic spikes like above, helping you to see what activities you need to replicate or capitalize on to grow your fan base. Btw, we discovered a few bugs after we released this feature so you may notice a few discrepancies with your data, but hang tight as we fix them as soon as we can.