Ticketing the Edinburgh Fringe

I’ve been going to the Edinburgh Fringe for many years.  It’s the world’s largest arts festival.  This year, 3,893 different shows were performed in over 300 venues around the City.  That led to sales of more than 2.6 million tickets.  In terms of ticket sales, it’s the third largest ticketing operation in the world, beaten only by the Olympics and the World Cup.  Unlike the Olympics and the World Cup, the Edinburgh Fringe takes place every year, so the pressure on the ticketing system is immense.

In many ways, the Fringe, in its current form, only exists because of the ticketing system run by the Edinburgh Fringe Society.  They publish the master programme guide and run EdFringe.com, which provides the single point of information and purchase for most attendees.  It’s a complex job, as it needs to coordinate with other ticketing systems run by larger venues.  Ensuring that the last remaining ticket for any show is not sold multiple times is a complex challenge in such a distributed system.  Which means that it is vital that it works.

The Fringe has come close to disaster in the past.  In 2008, the Fringe Society  introduced a new ticketing system, which failed on the opening day of ticket sales.  It should have been a “never again” lesson, but this year, when I went to book tickets, there was an unpleasant reminder that the lesson may not have been learnt.

The 2008  disaster was largely saved because tickets were still paper and most people bought them at the Fringe Office.  Today, almost all are bought online.  The easier it is to purchase them, the more people are likely to buy.  So, when you’re designing the user interface of a ticketing app, you should attempt to minimise the number of user interactions to buy those tickets.  For the Fringe, it should look something like this:

The first step is to choose your show, which means typing it in, getting a drop-down list as you go, and then clicking on the one you want to see.

That should take you directly to the show information.  The EdFringe website shows you the key information – the show details, the venue, the time and duration of the show and the dates when tickets are available.  To book tickets with a well-designed interface, you’d just click on the day when you want to see the show:

That click would take you to the order page, where you’d need to make a decision about how many tickets you want (as well as which type of concessions, if applicable).  The default should be one ticket at full price, so you just need to add it to your basket.

The only thing left is a confirmation screen, with the options of paying, editing your choice, cancelling, or booking more tickets.  If you choose to book more, it should pop the ticket in your basket and return you to the search screen. 

That’s how it should work.  After typing in the first few letters of the show name, it needs just four clicks to add it to your basket.  When I tried booking tickets this year using the  EdFringe ticketing system I found it needed an unbelievable sixteen clicks to get through that process.

There’s a rule of thumb in web design, which is that each additional click loses about 50% of users.  That’s probably not going to happen with a ticketing system, as users are there because they want to buy tickets and will persevere.  But it raises the question of the competence of the Fringe Society in their web procurement.  They got it disastrously wrong in 2008 and it feels that they may be heading down the same road again.

There’s another equally annoying problem with the current system.  After you’ve popped around a dozen tickets in your basket, it starts to become unstable.  I spent around an hour selecting just over 50 tickets, at which point it cleared my basket.  Rather than risking another wasted hour, I booked them by phone – something I’d not done for at least ten years.  That took about 75 minutes, which probably cost the Fringe Office around £40 in staffing costs.  So, there is a tangible downside to getting it wrong.

A cynic might claim that’s an upside for the Fringe Society.  Many people I spoke to in Edinburgh this year expressed their frustration with the system and said that they’d only booked five or six shows at a time, because of the difficulty of using the ticketing app for   more than that.  The problem with that approach is that the Fringe charges a £1.50 booking fee per ticket, with a maximum of £7.50 per transaction.  If you split your ticket order into multiple transactions, you end up paying multiple £7.50 fees. 

This year, there has been more media coverage than usual about of the costs of bringing a show to the Fringe.  Producers and performers need to get audiences to make shows viable, which means that there should not be any obstacles put in the way, such as a badly designed ticketing system.  Yet at the Fringe Society AGM last week, the issue exercising the Board of Directors was not ticketing, but how to distribute grants to performers.  That continues a long-established tradition within these meetings of totally ignoring audiences, and electing Board members whose desire to sit on the Board generally seems to be about vested interest rather than the continued health and good governance of the Fringe.  An independent review of the 2008 ticketing fiasco highlighted this lack of governance, but I wonder whether anything has changed.

One thing that has changed is that the Fringe has a new CEO in Tony Lankester.  Whatever happened this year will have predated him, but he needs to be very aware that the Fringe Society’s primary purpose is to provide a platform to attract audiences and sell shows.  I’ve been told that this year’s ticketing system is an interim step to a better designed web experience.  To that end, it was good to see a researcher outside the Fringe Office soliciting user feedback.  But that doesn’t alter the fact that this year’s website sucks and nobody seems to be talking about it. 

At the end of this year’s Fringe, the Stage gleefully reported that ticket sales have stalled, with sales down from 2,612,913 in 2024 to 2,604,404 this year.  Six and a half thousand is hardly a major fall, and probably can’t be ascribed to a bad web interface.  However, the ongoing concerns about funding shows means that the future of the Fringe needs to find a way to increase ticket sales.  There are plenty of issues for producers and performers, not least the cost of accommodation and a changing audience demographic.  However, having a plan to make ticket purchase easy should be top of the list for the Fringe Society, as it’s one of the few things they can influence.  Audiences are critical to the Fringe’s future.  Yet it’s telling that in a recent article about their five-year plan, there’s not a single mention of the “a” word.  Their development goals are Thriving artists, Fair work, Climate action, Equitable Fringe, Good citizenship and Digital evolution, but nothing about audience growth.  That needs to be addressed.

The only thing that can be said about the current ticketing systems is that it would make a good teaching resource of how not to design a web based ticketing system.  I’ve shared my analysis of that with the Fringe Society.  If anyone would like a copy, let me know.