Monday, 31 October 2011

Abundance and Scarcity and Agencies

So, if you divide up "things you sell tickets for" into "things that will easily sell out" and "things that aren't going to sell out", and accept that you need to treat them differently... where do agencies fit in?

It should be fairly clear that a ticket agency can be very helpful with a show that isn't going to sell out. They can be given an allocation, and that doesn't matter much because there is less competition for the other tickets. They can provide additional marketing, and they have a list of customers which spreads across venues and promoters, so they can reach customers you otherwise wouldn't be able to reach. They can offer tickets to shows across a geographical area and become a portal, so that the casual visitor can see what's on and be directed towards tickets without having to check each venue. They add value, from the promoter's point of view, and it would be reasonable to give them a commission on the tickets that they sell that wouldn't otherwise get sold.

It should be equally clear that a ticket agency doesn't have much to add with a sell out show. If it's popular, has an established fan base, has rave reviews, people will know about it anyway, without the benefit of marketing. Adding an extra layer to the sales process makes things less efficient and more expensive. And should anything go wrong, necessitating a refund or a cancellation or an exchange, the promoter can't do anything about it because the agency took the money, and the agency won't do anything about it because they don't care. If an agency gets an allocation to a show that would sell out anyway, then once all the tickets remaining at the box office have been sold, the agency will have a monopoly on the remaining tickets, and commission notwithstanding can charge whatever booking fee they like on top of the "face value" to the people scrabbling for the last few tickets. They make life harder, from the promoter's point of view, but can make a lot more money a lot more easily from the agency's point of view.

There are a few agencies that specialise in "shows that won't sell out" - the TKTS booth is the first that springs to mind. And there are many many more that make a living out of touting those last few tickets to sold out shows, adding nothing of value and promoting the impression that "booking fees are a rip off" - because in this case, they are.

Abundance and Scarcity

There are all sorts of niches and divisions within "Ticketing"; ticketing for Theatre is very different from Rock and Pop is very different from Sports is very different from Attractions, and so on.

But there's another division that cuts through these different markets, and fundamentally affects everything about the process.

1. If your tickets are going to sell out

Then what you care about is the efficiency of the system. You don't need to do much marketing, you don't need to do fancy discounts, you don't even much care about customer service. If any individual customer has a bad experience, it doesn't matter, there are plenty more. Humans are hardwired to value scarcity; if your customers know that tickets are in short supply, they will set alarms for first thing in the morning, they will queue up overnight, they will pre-register, they will pay through the nose, they will put up with all sorts of shit in order to get their tickets.

You care about fraud. You care about ticket touts. If the tickets are scarce, then it's almost certain that their street value is higher than their "face value" (see: stupidity of concept of "face value", and I'll be writing about the after market later).

As long as you can sell all the tickets without mass disruption (i.e. the website or phones crashing), you can process all the money, and you can minimise the overhead, you're happy.

2. If your tickets are not going to sell out

Then performance is the bottom of your list of problems. Fraud is irrelevant. Ticket Touts can't make any money off of your, so will leave you alone.

Having a really pleasant to use sales interface on the website is important, so people don't get frustrated and go away. Having well targeted marketing, so you reach the right people who might be interested, but not so often and so indiscriminately that they unsubscribe is important. Special offers to encourage people who are already coming to bring their friends. Social network integration to bring people's friends along. Membership schemes and discounts. CRM and good customer service to keep people happy and keep them coming back.

And the thing about these two different modes of operation is that you'll get both of them in any venue, so any system has to cater for both, but they have competing requirements. The streamlining that you need for the best handling of sellout shows doesn't allow for the complexity you need for maximising sales to non-sold-out shows.

Wednesday, 26 October 2011

Why is ticketing hard?

Everyone has to deal with the hard limitation of supply. If you oversell a book, or you run out of jeans or mars bars or whatever, you can just order / print / make more. With tickets, although very occasionally an extra performance is scheduled in response to demand, if you sell out, that's it. Which is why everyone hits the phones and the website at 9:00:01 when the tickets went on sale at 9:00:00, and it also means that speed of transaction processing is limited by needing to keep a completely accurate count of exactly how many tickets are left. Even if you've a million tickets to sell, you still need to know precisely how many remain before you let someone put one in their basket.

If you ever do anything popular, you will have to content with the spikiness of demand. Unless you're JK Rowling, you're unlikely to cause such a burst of traffic when your book goes on sale that you affect Amazon's servers in any way. Take That can still crash the TicketMaster site. Most things that aren't tickets sell at a relatively reasonable rate, spread out through the day. When hot tickets go on sale, EVERYONE hits the site at once. The number of servers and telephone operators you need to handle peak load is thousands of times that which you need to handle average load, and the rest of the time they sit round doing nothing and costing money. And this spike of demand is caused in part by the second factor...

If you have assigned seating, it matters which ticket you have. Whilst you are being hammered by a massive load of people all hitting the site at once, and keeping track of the remaining stock, you also need to make sure that you sell seats that are next to each other, and that your customer is getting seats B14, B15 and B16, and not B14, B15 and C20, because someone else nipped in to by B16 from under you whilst you were in the middle of processing the transaction. This effectively forces you to process requests one at a time (within each part of house), without any of the concurrency that would usually give you better performance.

Once you've got over the simple (!) stock control problem, you have to consider the pricing. Which is usually ridiculously Byzantine. There will be price breaks in the auditorium, different prices for matinees, student and OAP and child concessions, early bird discounts, members schemes, special offers, promotional codes, coach ticket rates, booking fees, per ticket fees, transaction fees, delivery fees, and anything else that a venue can invent, usually in the name of "promoting accessibility" but with the usual result of making sites run slower and the lives of box office staff harder.

Booking fees

Everyone hates them; why do they exist?

Tickets are sold in a really odd way. When you go to Tesco's, you don't see your groceries labelled on the shelf with the cost price, and when you take them to the checkout suddenly discover that you also have to pay a shelf stocking fee, and a credit card processing fee, and a head office administration fee. The cost of sale is included in the displayed price; it is assumed there is a cost price that Tesco's paid, and they're selling it on for a profit.

From the point of view of the box office, or the ticket agency, the "Face Value" of the ticket is the cost price. The price that's advertised to the consumer is the amount that the producer expects to get for the ticket - there's no margin to cover the cost of sale. Which is why the whole cost of running the ticketing operation - call centres, staff, webservers, developers, licences - has to be covered out of a "booking fee" lumped on top of the ticket price.

It wasn't always like this. When I was writing the "Deal Calculator" for Artifax Event - a nice little bit of functionality that let you put in the projected or actual box office take, and worked out the split and guarantees and first calls and so on, and reported how much money went to the venue and how much went to the promoter, there was a field for box office commission, that had a default value of 10%. Normally, the gross take would have the box office commission, credit card processing fees, PRS fees, taxes and so on deducted, and then what was left would be divvied up between the promoter and the venue.

What changed this, to grossly over-simplify, was TicketMaster. They'd do a deal with the venue or promoter and offer to take all the tickets and sell them without an inside commission, which meant that the venue or promoter could completely avoid the cost of sale. Brilliant; an offer you can't refuse. TicketMaster would then have a monopoly on the tickets for that event, and if it were a big, definitely-going-to-sell-out must-see show, they could charge what they liked on top as the "booking fee", and the punters would pay it. Which is why if I look up X Factor 2012 tickets right now on the Manchester Evening News Arena website it lists them as £32.50, and if I click on the "buy now" button I go straight through to the TicketMaster site, where they are listed as £38.00: £32.50 + £5.50 in fees. That's about 17% in fees. Plus a 2 pound delivery fee, which you pay whether you collect the tickets from the box office, have them posted to you, or even choose to receive them by email, so if you're ordering two tickets, that's a 20% markup, or twice the traditional box office commission.

Now, don't get me wrong. Selling tickets is complicated. 20% is not an unreasonable markup for a retail operation. Amazon takes about 20% on books, for example, and whilst they've got to worry about moving physical lumps of preprinted paper around, with ticketing, you've got your usual costs of call centres and webservers and delivery and fulfillment, but in addition to that there are unique factors that make ticketing hard, which I will expand in a later post.

So it's not the size of the booking fee that's the problem: it's the fact that it's exposed to the customer. Those X Factor tickets should have been advertised as costing £39, and everyone would know what they were buying, and the retailer or box office or agent should have a margin on the inside of that price where they make their profit and pay their costs.

The distinction between "face value" and "booking fee" is enshrined, if not in law (I just had a quick look, I'll google harder later) then at least in the Code of Practice of the Society of Ticket Agents and Resellers:

They define

5.  The “face value price” of a ticket means the
price recommended by the organiser of the
relevant entertainment or event as the price at
which the ticket would normally be sold direct
to consumers by the Box Office, including any
mandatory additions imposed by the organiser
such as a “Restoration Levy”. This price will
be displayed on the face of the ticket itself and
therefore excludes any mark-up or additional
charges that may be levied, and excludes any
discount or other concession

which sounds reasonable until you consider that in many cases the agency has an exclusive deal, there is no box office from which you could buy tickets to avoid the booking fee, the box offices themselves charge booking fees these days, and that the public are perfectly capable of comparison shopping for basically everything else, so there's no reason for tickets to be different, and presenting the prices differently in fact makes matters worse.

Monday, 17 October 2011

What is a ticket?

Sounds like a stupid question, I know, but most of the unfixable problems I've run into with all the other ticketing systems I've worked with have come from failing to answer this question.

Consider the most basic, primitive ticketing system imaginable: a book of raffle tickets. There are two basic interactions with this "system":
  • As a customer, I go up to the box office and hand over some money, and in return receive a piece of paper. 
  • As a customer, I go up to the entrance to the venue, and hand over my piece of paper to an usher, who tears it in half and allows me through the door. 
In this example, a single, humble piece of paper is fulfilling two entirely distinct functions:
  • Firstly, that of stock control. The main purpose of a ticketing system is to prevent you from overselling the event, from breaching the fire regulations, from selling the same seat twice. If there are no raffle tickets left in the book, you cannot sell any more. 
  • Secondly, access control. The main purpose of a ticketing system is to prevent people who have not paid the fee from attending the event (“event” here may include performance, train journey, museum exhibit, etc).
They're both the "main" purpose, because they're both encapsulated in the idea of "selling tickets"; it's the idea of a ticket that's confused between these two functions.

When all you've got is a hammer, everything looks like a nail... and right up to the invention of the computerised box office system, the convenience of using the same tokens that you use to keep track of sales volumes as the token that lets you through the door was irresistible.And so the terminology and business process of box officer were based around these bits of paper, and the first box office systems, specified by box office managers, embedded these ideas in their architecture.

Trouble is, this all gets terribly confusing when you start selling - and then delivering - tickets on the internet. An efficient ticketing system for the internet age must be clear on when you're dealing with "the right to occupy a seat" - the stock control, when you're dealing with "the unique serial number or bar code that you assign to the ticket so you can check that it's valid" - which I will hereafter refer to as "the access control token", and when you're dealing with the physical medium that's carrying the access control token - perhaps a piece of card from a BOCA printer, perhaps a sheet of paper from the home inkject printer with a barcode on it, perhaps an RFID card or a mobile phone or any other device - which I no longer have a good name for, because "ticket" is a much to loaded term to use, and "print out" doesn't cover the act of "loading some data onto a smart card".

Yet Another Ticketing System

I left my last job, as Director of System Development at Galathea STS Ltd, the producers of the enta ticketing system, in May 2009. Since then I've been writing a brand new ticketing system from scratch; it seemed like it would be easier to write a whole new ticketing system, by myself, with no backup or staff or resources, than it would have been to turn enta into the kind of system I could be proud of.
Over the next few months I plan to explain why I think there's room in the world for another ticketing system, and why I believe that the system I've built was worth writing.