Understanding customers beyond purchases: clicks, keywords and browsing behavior

March 25, 2009 2 Comments

Traditional CRM efforts used to focus on customer’s purchasing behavior. In the last decade, customers shifted their behavior online. The amount of data captured by online systems are quite large, and often left in separate systems or tools (Coremetrics, Webtrends, Email service providers).

This data has tremendous value.

1) It allows marketers to better predict customer’s future purchasing probability. We’ve recently built a model using traditional purchasing related variables and were able to have a gain curve where top 40% of people ranked generated 87% of revenue. When we incorporated email open, click and email status variables into the mix, our predictive power increased from 87% to 94% for the same top four deciles.

2) It allows marketers to find the next best offer or next best product to offer.

Integrating email related data and web-browsing data is crucial in improving targeting. This will not only improve the top line, but also reduce the unnecessary spend due to reduction in offline spending.

Ömer Artun
Managing Director | Agilone
www.agilone.com

Filed under Predictive Modeling Tagged with , , ,

2 Responses to Understanding customers beyond purchases: clicks, keywords and browsing behavior

  1. Pingback: Predictive Modeling Genius « Sound Loyalty

  2. Colin Brogan says:
    March 29, 2009 at 8:12 am

    It’s truly amazing what one can do with large amounts of customer data. Certainly, these methods can increase the value of your current customer base, but i wonder what can be done on the acquisition side. If these frequent purchasers are truly “advocates” of the company, could you focus marketing spend on these folks to generate referrals, and leverage their ideas to improve the product/service bundle? In other words, are these customers “Promoters”, and if so, how can we harness them in other ways to grow the business?

    Reply

Leave a comment