Off to Bucharest, and "Competing on Analytics"

I'm just sitting in the departure lounge at Heathrow Terminal 2, waiting for my flight to Bucharest. I'm running the last of the Enterprise BI Masterclass seminars (at least in Europe) over the next couple of days, and this is a particularly interesting one for me as it's my first time in Romania. Whilst I normally stay in the hotel most evenings when I'm running these events, this time I'm going to try and get out in the early evening and try and see some of the sights, particularly the Palace of the Parliament (a.k.a. "Ceauşescu's Palace"), second only to the Pentagon as the largest free-standing building in the world. I'll post some photos if I manage to get out to see it.

I'm going to away a fair bit over the next couple of weeks, as I'm flying back from Bucharest on the Saturday morning and then flying out to Oslo on Sunday afternoon for the Norwegian User Group Annual Conference. I've been invited along to the event to do a couple of presentations, take part in the panel and so on, and interestingly it's being held on a cruise liner that sails around the fjords near Oslo. Again, I'll try and get some photos and report back on the sessions (at least the ones in English...)

Apart from the traveling, I'm currently working my way through a very interesting book, "Competing on Analytics" by Thomas N. Davenport and Jeanne G. Harris. If you're in to BI you may well have heard of this book already, it's pretty topical at the moment as it's central premise is that you can improve the performance of your organization by taking an analytical approach to strategy and decision-making. In a way (if you're into the Oracle database side of things as well) it's a bit like "Optimizing Oracle Performance" for business intelligence, in that (a) it advocates improving company performance (as opposed to application performance) based on statistics, diagnostic data and an analytical approach, and (b) it inspires a near-religious fervour in it's advocates, because it offers a straightforward, predictable and rational approach to performance improvement.

From what I've read so far, it's a very good book. Two of the early examples it uses to illustrate an analytical approach to performance management are Netflix, who use their knowledge of customer movie ratings and order clusters to make more effective movie recommendations, and Bolton Wanderers (or more accurately, Sam Allardyce) who used analytics to work out what playing strategy to use, which players to sign and which players to play on the day. Like all these books around a single idea, you could probably sum it all up in one magazine article rather than requiring a whole book (and in fact there is a magazine article, orderable here from Harvard Business Review, that sums it all up in just 12 pages) but so far it's fairly readable, a bit like Optimizing Oracle Performance before it got into all the queuing theory stuff.

Anyway, if I come up with anything else particularly interesting from the book I'll try and summarize it here. Other than that, the flight leaves in about an hour, time to check all the seminar demos are still working and I've got the address for the hotel once I land.