It’s still day 1 at the Future of Music Coalition Policy Summit 2011, and I’m sitting in the conference hall as I write this. Talking to Glenn from ReverbNation, and listening to this morning’s panel on ticketing has me thinking about software.
Specifically, I’m thinking about how text analytics (machines that can read and understand language) can circumvent Ticketmaster and others hoarding ticket-buyer data.
The Ticketing Panel
The ticketing panel had managers and promoters on it, and basically everyone jumped up and down on scalpers and TicketMaster (who deserve it), and put forward their different reactions to the messed up economics of the ticketing market. Some of them ditched ticket master for competitors, some of them set particular terms in various deals. Some people wanted to require the credit card used to buy the ticket presented along with a photo ID to get into the venue (which is, as one panelist pointed out, more security than you get for a plane – that’s just a photo ID).
In response to scalpers, one manager sensibly pointed out that if there’s an unfed demand for the show that drives up the prices way higher than face value one of two things is happening: 1) your tickets were too cheap to begin with or 2) you aren’t doing enough shows. That’s why scalpers are seeing a chance to sell tickets at higher prices. That’s why they buy them. If you raise the prices to what the market will bear the scalper won’t be able to drive up the price through artificial scarcity, and if you play enough shows, there won’t be scarcity at all.
There are difficult issues with where to play, how much to charge, whether to shoot for a quick sell-out or a high-end smaller audience, or a mix (maybe involving letting something of a secondary marke develop). There’s a whole side discussion to do with whether and how to participate as an artist in the secondary market with the scalpers, and dynamic pricing and all of that.
So, where does text analytics come in?
Right now a lot of show booking is a guess. The data we have about how many people bought tickets last time, how many records we sold in that market, etc., are not super detailed or super up-to-date. And a lot of times either the venues who sell tickets or present shows, or the dreaded TicketMaster, hoard the customer sales info, making it hard to compare with your mailing list and hard to study to make booking decisions or sell tickets.
Here comes software!
Customer management databases are nothing new. I don’t know exactly how well these things work, because I don’t use them, but I’m guessing they’re missing some demographic modeling and I know they’re missing text analytics. And text analytics can help get around Ticketmaster and venues who hoard customer data.
Let’s say I played a show with my band, the Octopus Revolution, in Toledo. My fan email list has a Sue Baharnd living in Toledo on it. I emailed her about the show, but I don’t know if she came out. TicketMaster won’t tell me. But @SueBaharnd tweeted a mention of my band and how awesome we are on the day of the show in her market. If I’ve got live text analytics in my customer management database I can use that to fill in the gaps that Ticketmaster is refusing to fill.
If she uses my Facebook app and invited some of her friends to an event she created for my show then I can figure out if she’s responsible for multiple tickets. If she checks in on Foursquare while she’s at the concert, that just makes life really easy. Having well integrated text analytics in my customer management database means that all of this can happen automatically.
Sounds a lot easier than fighting with Ticketmaster, doesn’t it?
Beyond missing data
There’s a lot more potential for automating previously unautomatable aspects of the artist-fan relationship. Does Sue need babysitting? Does Sue travel to New York a lot and would she like to see a show in Brooklyn while she’s there? Does Sue have a Tumblr blog with 15,000 readers? Would she write up your next album if she got it early? Does she hate e-mail marketing and love the telephone or facebook? Does she tweet about how great that weird ballad you slotted in at track 8 was, when track 3 was the hit song? There’s a lot of data out there about how exactly your customer wants to relate to you. And machines can read it for you.
The arts and entertainment are particularly fertile ground for this technology, by comparison with other brands. I don’t tweet that much about my shoes. And when I do, I don’t give out enough information for any machine to figure out the brand. But I do use social media to plan my leisure time and to discuss entertainment, and bands I do refer to by name.
Next steps
Obviously I can’t do any of this myself. But text analytics is coming a long way fairly quickly. Those of you out there who are working in text analytics – could this work? How long until it’s possible? 1 year, 5? Is it really a database input problem masquerading as a text analytics problem?
Those of you who work with customer management databases, particularly those of you do wish you had some more customer data from Ticketmaster or venues, does this seem like a feature you would like? Do you think it would be worth the upfront cost in software development to so radically improve your understanding of your fan base?