June Dershewitz on Web Analytics

The Best of eMetrics DC 2008

I have recently returned from a highly informative, thoroughly enjoyable and action packed week at the eMetrics Marketing Optimization Summit 2008 in Washington DC. Since then people have been asking me, "What did you think? How did it go? What did you learn?" so here it is in writing.

Big Successes

eMetrics would not be complete without a sampling of client-side case studies, and I knew that I would catch some at this conference. However, I was not expecting the case studies to be quite so inspiring and so motivating as they turned out to be.  I had two favorites, both rightfully featured on the big stage as keynotes: James G. Robinson from the New York Times, and Joe Megibow from Hotels.com.

James from NYT explained how the work he's doing in the online channel has influenced the offline channel at his company, and how his team has managed to convey the importance of web measurement to that wider audience.  For instance, they found that they could use daily page views (yes, page views) on their web site to accurately predict the number of print newspapers they'd sell the following day. Pretty simple, yet a very effective way to encourage cooperation across diverse groups and spread the word about web analytics.

Joe from Hotels.com described how his team has managed to fuse customer feedback and session-level customer experience data to identify and correct commerce flow problems on the web site. He gave great examples, and it's something that anyone with an online form (and isn't that everyone?) really ought to be doing at their own company. Joe also highlighted the work of his colleague Vernon, in the audience, as an ace developer of applications that the analytics team has been building in order to monitor their systems and notify them of problems. Now everyone wants to hire Vernon, or some rare individual just like him.

I hope that someone (ahem) will post videos of the keynotes online. These stories need to be shared.

Getting Executives to Listen

As a theme I've seen evolving over the past couple of years, web analytics professionals are working hard to get buy-in from our executive teams. At this conference there were a couple of really articulate and important presentations on this topic. 

First, all-around great guy Eric T. Peterson gave a presentation entitled Competing on Web Analytics.  It's an extension of Tom Davenport's 2007 book, Competing on Analytics, applied specifically to the online channel. Since his time as an industry analyst I've always felt that Eric has had a clear view of how the practice of web analytics can evolve, progress and grow within a company, and this latest presentation continues to build on that theme.   

Also, Liz Miller from the CMO Council gave a compelling presentation on The View from the CMO's Office.  I hope she returns to eMetrics, because she's got an important message we should all be listening to as we attempt to capture the attention of our CMOs.

Talking to Everyone

Those who know me know that I love to meet people, and I always stay up late in the Lobby Bar, talking shop while eating fine Belgian chocolates. I recognize that networking is one of the primary reasons that people attend conferences, and there were plenty of opportunities to do so at eMetrics (not just in the Lobby Bar). 

My personal favorite networking event, Web Analytics Wednesday, was as lively and well-attended as always - thank you Coremetrics, WebTrends and eMetrics for generously sponsoring us. After years of rogue and covert operations we were honored to have WAW make it onto the official eMetrics agenda for the first time.

Also, as a member of the Web Analytics Association Board of Directors I had a hand in organizing the WAA Community Meeting and Speed Networking session in DC. Based on feedback we'd gotten from our members we decided to help everyone meet each other by conducting this speed networking experiment - it was loud and chaotic and we might make some refinements next time but I'd still call it a success.  Here's a picture from speed networking:

 

That's Janet Park (WAA Membership Committee volunteer extraordinaire) on the left, introducing herself to June Li (my doppelganger, the "other" June of web analytics) on the right.

If you went to eMetrics DC, lucky you - I'm sure that you left feeling as inspired as I did. If you missed out, I do hope to see you at one of the 2009 events.

Book Review: Click by Bill Tancer

I love data. No, no, I mean it. When I saw Bill Tancer speak at a conference a little over a year ago he closed with that sentence, in fact a clever plug for his corporate blog: ilovedata.com. I gladly took the bait. Recently, when I heard that Tancer had published a book about his beloved data, I snapped that up, too.

The book is Click: What Millions of People Are Doing Online and Why it Matters. It is, chapter by chapter, a collection of analyses focused on a single data set, namely the search and clickstream behavior of 10 million US-based internet users. 

On the surface Click has genuine pop culture appeal, with coverage of topics like porn and fashion and phobias and rock bands. In that respect it's readable by anyone who uses the internet and wonders what we - the collective "we" - are doing online. At the same time I believe there's extra value for those of us who work with web activity data for a living. 

As an analyst there is nothing quite so satisfying as taking some data, asking and refining and asking questions, turning the data over and over until interesting observations pop out. Tancer writes about his observations, but he also writes about the processes leading up to his observations. In no way is Click meant to be an instruction manual for data analysts, and yet it does a pretty good job of describing the act of analysis. If you want to know what it's like to be a data analyst, read this book.

Some reviews I've read assert that Click is a plug for Hitwise and its own slice of competitive intelligence data. I disagree. Taking a step back, I think that smart analysts can draw from the book and relate it to whatever data they happen to have at hand. In particular, site-side web analytics data from tools like Omniture and Google Analytics, while offering a fundamentally different view of the internet, can still be used effectively to uncover trends in search behavior a la Tancer's approach. Simply ask and refine and ask questions, then present your observations with the same enthusiasm that Tancer does in Click.

Now that I've written this review I can hand off the book to my fellow Semphonic data analysts.  There's a waiting list. We all love data.