The Analytics 2011 Conference, taking place Oct. 24-25 in Orlando, FL, is quickly approaching. Larry Seligman, a session speaker at the conference, recently took the time to answer some questions about his presentation and what he is looking forward to at the conference. Larry is the Director of BI and Analytics for InterContinental Hotels Group. He is responsible for advanced analytics related to campaign optimization, customer scoring, performance analysis and the IT analytics agenda. The title of his talk is “Your Customers Don’t Care About R-Squared.” Below he gives us some additional information about his talk and analytical trends in the hospitality industry.
How long have you worked in the hospitality industry?
Five years.
Can you give us some examples of the types of metrics you will discuss in your presentation?
Model fit metrics such as AIC and r-squared, and parameter metrics such as Wald Chi-Square t-statistics are meaningful to statisticians but meaningless to most stakeholders. The stakeholders have different metrics such as ROI, response rate and incremental revenue. My presentation is on bridging the gap between statisticians and stakeholders by showing a path from statistical metrics to business metrics.
How have these metrics affected the way IHG makes decisions?
In some cases, the entire customer selection, offer selection, performance forecasting and reporting duties have been given to our Predictive Analytics team. We’re answering questions about whether or not certain campaigns are worth running, and how much budget should be put toward a given campaign.
In your opinion, how is analytics changing the hospitality industry?
The hospitality industry is trying to personalize and customize, and is very focused on the guest experience. We can’t do that without sufficient data and actionable analytics. But if I want to give you the hotel room that you want with the tweaks and perks that you want, then I need to know what you want and what the value proposition for providing it to you is. Analytics can provide all that.
What do you see as the hottest trend in analytics?
Big data and social media are probably the hottest things, but for a lot of companies they aren’t the most practical things. There is a virtuous cycle of good data leading to actionable rewarding analytics that can then lead to funding for the capture of more good data. There is also the vicious version of that cycle. Leaders need to understand where their processes are in terms of those cycles, and where there is value in investment. That understanding starts with cleaning up and understanding their data. Understanding their data is the most practical thing that many organizations can do right now, especially since there are so many new types of data that are now available.
Have you ever attended Analytics 2011 (formerly M2010 and F2010) before?
No, but I’m really looking forward to it. I’m grateful to SAS for the opportunity to present.
Why are conferences such as this one important to the analytics community?
Analytics beyond simple reporting is quickly evolving, and is still very young. The domains for applying analytics, like big data, are so new and so rapidly evolving that many companies are inventing methods for analyzing the data. So, many of us across many companies are simultaneously trying to build similar wheels. We should get together at conferences like Analytics 2011 so that we can share learnings and try to keep pace with all this change. 
