CTS Eventim finds Exasol just the ticket for event recommendations


The data science team at Europe’s largest ticket vendor has used the analytical database to build a system that keeps concert goers and rock fans up-to-date on shows coming soon, to nearby venues.

Eventim gigDuring 2017, Europe’s largest ticket vendor CTS Eventim sold more than 250 million tickets, the company reported in late March. That total includes around 150 million tickets for more than 240,000 live events – from rock concerts to ice shows – as well as some 100 million cinema tickets.

And, for the first time, the company’s annual revenues exceeded the €1 billion mark, to reach €1.034 billion, from the tickets the company sells and the events it stages in 25 countries worldwide.

As Eventim continues to grow its online operations, expand internationally and launch new events such as Germany’s New Horizons dance music festival, which took place for the first time in 2017 and is scheduled again for August this year, customer experience will continue to be a top priority for its data science team, says Fred Tuerling, its Senior Vice President of Information Services.

Most important, he adds, will be steering customers in the direction of events that will appeal to them, held at venues that are convenient:

This, for us, is a fascinating data challenge. There are really brilliant events taking place all the time, but people these days are so bombarded with advertising and information that they do not always notice that there is a show, concert or other event due to take place, at a convenient location, on a date when they’re free. One thing we hear from many customers is this: ‘If I’d known about it, I would have definitely bought a ticket.’ We want to be sure that they don’t miss out on fantastic opportunities to be entertained.

When CTS Eventim first established its data science team almost four years ago, he adds, tackling this challenge appeared high on the list of priorities very early on. In particular, the team recognized that an effective recommendation engine delivering relevant, appealing events suggestions to customers would enable information services to become a revenue generator, bringing incremental business to Eventim, rather than a cost center.

But in order to achieve its goal of becoming a data-driven revenue-generator, the team also knew that it needed a data hub on which to consolidate information from a very wide range of systems and sources, says Tuerling. Some of that information is business-to-business data; for example, data relating to 6,000 promoters who create and execute events. A great deal more is business-to-consumer data, relating to 17 million consumers who have bought tickets from Eventim in the past and may well buy again in future.

Analytics choice

For this hub, Eventim chose Exasol’s analytics database, in part because a number of team members had used it in previous jobs. For example, Dr Ulrich Fricke, head of information management and technology at Eventim, had used Exasol during his time working at Xing, the German social networking platform.

Because Exasol is in-memory, columnar and does massively parallel processing (meaning it’s partitioned across multiple servers with each one having memory/processors to process data locally and concurrently), Exasol scores very highly on performance in benchmarking tests, such as those run by the Transaction Processing Council (TPC).

The data sciences team at Eventim has used Exasol as the platform for its homegrown CRM system, combined with a recommendation engine. According to Tuerling, this system has supported the sale of several hundred thousand tickets last year through recommendations. The incremental revenue associated with those add-on sales, he reckons, has enabled the system to pay for itself, but just as important, it has helped Eventim get to know its customers better, which will pay further dividends over time:

What we do is create individual user profiles. We start with you as a customer and all the data you’ve shared with us, starting from simple information such as your address to tickets you’ve already bought. That already tells us a great deal about your tastes and the venues you might attend. But we can go further still to get clues that help us make recommendations, including the events or shows that you’ve looked at on our site and also, where you’ve allowed us access, the artists who you listen to on iTunes, for example.

In this way, we are going beyond the standard approach of, ‘Customers who bought that also bought this.’ Our goal is to have a 360-degree view of the customer based on as many as 500 attributes relating to them, so that it’s truly personal to you and your tastes and your location. I see our work in data science, and in particular recommendations, as a service not just to the business, but also to customers. There are so many events, in so many places, that it can be hard to find the right one, so customers genuinely value and appreciate recommendations that keep them up-to-date on what’s happening.

Image credit - Pinterest

    Leave a Reply

    Your email address will not be published. Required fields are marked *