Lessons From the Emetrics Summit

The state of the art of web analytics advances

Over the last three years, the focus of the Emetrics Summit — an executive forum on measuring Web site success — has gone through three major phases. First, it simply sought to define "Web analytics," next it focused on instruction, and finally it has come to concentrate on solving business problems.

"This year the focus was markedly different," says Jim Sterne, CEO of Target Marketing and founder of the forum. "Vendors now have multiple methods of capturing customer data and we are starting to see how we can make all this information work for us."

Terry Lund, a Web site strategist in attendance, says, "Web analytics (better intelligence) can now deliver the relevant information to support stated business goals and objectives, but the key is to relate the Web metrics to the corresponding elements of the business value creation model." One area where Web analytics can give direct data about customer value, for example, is the recency and frequency of visits to a Web site.

Web Analytics ToolKit

Types of analysis you may need and a selection of relevant vendors

Clickstream analysis (with reporting)
Omniture, Webtrends, Coremetrics, ClickTracks, WebSide Story

Web site performance & navigational analysis
Keynote, Maxamine

Search engine optimization/marketing
BidRank, Inceptor, EC Next

Email campaign reporting & management
CheetahMail, Responsys, Constant Contact

Surveys & satisfaction
Forsee Results, Opinion Lab, Vividence

Clickstream and offline reporting integration
Look for an open API so you can get clickstream data into your CRM system

Netraker, Usability Sciences, a lot of boutique firms

Competitive intelligence/benchmarking
comScore, Hitwise

Ad management reporting

Brand protection/monitoring (RSS Feeds, Multimedia)
Cyveillance, BzzAgent

Forecasting/predictive modeling
SPSS, Intelligent Results

Enterprise search analytics
Endeca, iPhrase, Mondosoft, Verity

But not all of the news from Emetrics was about what's new. "What many businesses find surprising is that the 'old' ideas of direct marketing work just as well in this new era of online marketing," says Jim Novo, author of Drilling Down: Turning Customer Data Into Profits With a Spreadsheet (, 2001). "When [customers are] in control of [their] actions — as they are when they can take action any time they want to, 24 hours a day seven days a week — the old models used in direct marketing to predict response and customer value end up being even more accurate," says Novo. "As friction is reduced because they are more in control, their behavior becomes more predictable."

A customer behavior analysis expert since 1997, Novo was the former VP of programming and marketing at the Home Shopping Network. His system of measuring a customer's potential value by tracking the recency and frequency of visits — both easily accessible through Web logs — provides a way to calculate the potential value of visitors to the site.

Recency is defined as the time elapsed since the last visit, purchase, or action taken on the Web site and frequency is how often someone visits. The more recently a customer engaged in an action, the more likely that customer is to repeat that action. And customers who are more likely to repeat an action respond to marketing or service offers at a higher rate.

"If it becomes easier and less time consuming to accomplish the same task on the Web, they become more likely to follow through and complete the action," explains Novo. "These metrics are a way to track and manage these 'likelihood to act' ideas at the individual customer or customer segment level."

These metrics have been used successfully offline for decades, and they work even better when profiling online behavior. Because customers in different stages respond to offers differently, it also becomes a way to gauge how to segment marketing activities and email campaigns.

Sally Falkow is a Web content strategist and can be reached through

Editor's Choice
Brian T. Horowitz, Contributing Reporter
Samuel Greengard, Contributing Reporter
Nathan Eddy, Freelance Writer
Brandon Taylor, Digital Editorial Program Manager
Jessica Davis, Senior Editor
Cynthia Harvey, Freelance Journalist, InformationWeek
Sara Peters, Editor-in-Chief, InformationWeek / Network Computing