In my 10+ years at SAS, I’ve helped customers with challenges as diverse as forecasting flight traffic, an early-warning detection system for electronic payment grid events, fraud detection at the European Anti-Fraud League (OLAF), credit scoring in a Basel II context, as well as various customer related predictive modeling and marketing optimization projects across the Financial, Retail and Manufacturing sectors. In short, I love getting my head around a new business challenge and finding the right analytic approach to tackle it!

In de glazen bol:
Wat zijn volgens u de grote uitdagingen voor kwantitatieve marketing in de toekomst?

I’ve got two different answers for this: one general and the other a bit more specific. Generally, the greatest ongoing challenge I see for quantitative analysts is keeping the results digestible so that management stays on board. It’s no longer a question of ‘Will organizations make analytics a part of their competitive strategy?’, but ‘How soon will they start to reap the benefits of competing on analytics?’ Analysts have a responsibility to challenge the pervasive statistical trauma in our society by explaining in clear terms (and even clearer in a graphical representation) what the analyses are telling about our customers and how this will help us to define the best business strategies.

More specifically, I think that defining, measuring, and re-evaluating customer lifetime value is an emerging challenge in marketing analytics and will only continue to grow in importance. When choosing which customers to target for a particular campaign or special offer, it’s essential to have a good view on both the short and longer-term customer potential to make balanced strategic decisions. Too often, sales and marketing strategies put unbalanced emphasis on the next period, the next customer or the next sale. It can be equally important to take a step back from the customer’s current situation and objective evaluate long-term potential. Customer Life Time Value analysis can help to achieve this.

A of B vragen à la tête du client:
- Neural Networks or Decision Trees? Simplicity rules! A decision tree is easier to explain in pure business terms (eg: Customers who did this, and live here, are more likely to buy more of those) and thus better suited for shaping strategies (eg: Promote more of those, here) . But everything in its right place of course…A quickly-fitted Neural Network can give a first view on the potential in a new data set (but watch out for overfitting!!!!).
- Tennis or Squash? If you would have asked me 10 or even 5 years ago, Squash would have won hands-down. Better workout, more intense play, and it is incredible the amount of stress you burn by smacking the hell out of a little rubber ball. But as I get older, and the knees and back start to show some signs of wear, tennis plays more relaxed and it seems much easier to find a partner.
- Werchter or Pukkelpop? Even though (or perhaps, just because) Werchter always gets the biggest names, I prefer the more eclectic offerings and low-key atmosphere at Pukkelpop. If you understand why Faith No More is a better band than Metallica, then you’ll understand why Pukkelpop is better than Werchter.
- Half-full or Half-Empty? I tend to be an optimist in most situations. However, when it comes to sustainable food and energy and keeping the world hospitable for future generations, I definitely feel like the glass is almost empty and we all need to take socially-responsible actions now!