A Multichannel Forensics Case Study From MarketingSherpa And Paul Fredrick

This brief Marketing Sherpa case study describes the outcome of a Multichannel Forensics project at Paul Fredrick from last year.

Hint: You can mail fewer catalogs and be more profitable!

Marketing Sherpa Article: Fewer Catalogs, Same Results: 6 Lessons for Saving Big on Mailing Costs.

Click here to contact me for help with your Multichannel Forensics project!

The MineThatData E-Mail Analytics And Data Mining Challenge

It is time to find a few smart individuals in the world of e-mail analytics and data mining! And honestly, what follows is a dataset that you can manipulate using Excel pivot tables, so you don't have to be a data mining wizard, just be clever!

Here is a link to the MineThatData E-Mail Analytics And Data Mining Challenge dataset: The dataset is in .csv format, and is about the size of a typical mp3 file. I recommend saving the file to disk, then open the file (read only) in the software tool of your choice.

This dataset contains 64,000 customers who last purchased within twelve months. The customers were involved in an e-mail test.

• 1/3 were randomly chosen to receive an e-mail campaign featuring Mens merchandise.
• 1/3 were randomly chosen to receive an e-mail campaign featuring Womens merchandise.
• 1/3 were randomly chosen to not receive an e-mail campaign.

During a period of two weeks following the e-mail campaign, results were tracked. Your job is to tell the world if the Mens or Womens e-mail campaign was successful.

Historical customer attributes at your disposal include:

• Recency: Months since last purchase.
• History_Segment: Categorization of dollars spent in the past year.
• History: Actual dollar value spent in the past year.
• Mens: 1/0 indicator, 1 = customer purchased Mens merchandise in the past year.
• Womens: 1/0 indicator, 1 = customer purchased Womens merchandise in the past year.
• Zip_Code: Classifies zip code as Urban, Suburban, or Rural.
• Newbie: 1/0 indicator, 1 = New customer in the past twelve months.
• Channel: Describes the channels the customer purchased from in the past year.

Another variable describes the e-mail campaign the customer received:

• Segment
• Mens E-Mail
• Womens E-Mail
• No E-Mail

Finally, we have a series of variables describing activity in the two weeks following delivery of the e-mail campaign:

• Visit: 1/0 indicator, 1 = Customer visited website in the following two weeks.
• Conversion: 1/0 indicator, 1 = Customer purchased merchandise in the following two weeks.
• Spend: Actual dollars spent in the following two weeks.

Ok, that represents the basics.

By April 30, you are encouraged to write a paper that answers the following questions. Winning submissions will receive a copy of my book, Hillstrom's Multichannel Forensics, currently available at ForBetterBooks and Amazon.com. There's nothing wrong with winning a book valued at $95, is there??

I will give away at least one book, and as many as three books, depending upon entries within the following categories:

• The E-Mail Blogosphere: If we get enough entries, I will give away one book to the e-mail blogger who provides the most insightful answer.

• The Direct Marketing Industry: The best answer among direct marketing and e-mail marketing professionals and e-mail marketing vendors will receive a book. In addition, I'll publish well-written and insightful answers received from any qualified e-mail marketing vendor. In other words, you'll earn an opportunity to advertise for free to the MineThatData community, a community of more than 1,200 subscribers and daily visitors.

• The Data Mining Community: Data Mining professionals and University students are encouraged to send in entries, with the best-written and most insightful response receiving a free book.

Here are the questions you are encouraged to answer.

• Which e-mail campaign performed the best, the Mens version, or the Womens version?
  

• How much incremental sales per customer did the Mens version of the e-mail campaign drive? How much incremental sales per customer did the Womens version of the e-mail campaign drive?
  

• If you could only send an e-mail campaign to the best 10,000 customers, which customers would receive the e-mail campaign? Why?
  

• If you had to eliminate 10,000 customers from receiving an e-mail campaign, which customers would you suppress from the campaign? Why?
  

• Did the Mens version of the e-mail campaign perform different than the Womens version of the e-mail campaign, across various customer segments?

• Did the campaigns perform different when measured across different metrics, like Visitors, Conversion, and Total Spend?

• Did you observe any anomalies, or odd findings?
  

• Which audience would you target the Mens version to, and the Womens version to, given the results of the test? What data do you have to support your recommendation?

E-mail your responses to me by 11:59pm on Wednesday, April 30, 2008. Good luck, and have fun analyzing the information! Dazzle our readers with your insights --- feel free to share your findings in the comments section of this post.

MineThatData Traffic Analysis

Click on the image to enlarge it.

You're probably wondering how many folks read this blog, right?

Or, maybe the topic is only of interest to me.

Either way, you get to learn all about blog audience dynamics today.

According to my statistics, I have about 3,900 visitors per month (via Google Analytics) and an average of 450 daily subscribers (via Feedburner).

The image illustrates how we get to subscribers, and the true amount of monthly traffic generated by the blog.

During an average month, 1,873 visitors arrive via search (almost all via Google). 976 visitors arrive from other websites, about eighty percent from blogs referring folks to MineThatData (thank you all!). 200 visitors type in the URL.

Of these visits (3,049), about two percent (60) become what I call "Loyalists", folks who frequently visit the URL, or subscribe by RSS or E-Mail.

I estimate that each month, I add 60 subscribers, and lose 30 subscribers, yielding a net increase of 30 subscribers. To date, about 450 of you subscribe via RSS, E-Mail, or type in the URL frequently. You are my "Loyalists". Thank you!!!!

The 450 "Loyalists" drive the relative popularity of this site, visiting about 13,650 times a month. There are 854 actual site visits, 5,460 readings in Google Reader, 1,092 via E-Mail, and 7,098 readings in other RSS Readers.

In total, this yields 14,504 actual readings from my "Loyalists", and 3,019 visits from folks who weren't impressed with the site, or got what they needed from one visit and moved on.

In total, this yields 17,523 real "visits" each month.

There are flaws with the analysis --- there are probably many more "Loyalists" than measured by Feedburner --- BlogJuice suggests I have twice as many Bloglines subscribers as Feedburner suggests I have. So there might be as many as 1,000 actual "Loyalists", each visiting half as often as I have estimated.

Still, the number of real "visits" is a reasonably accurate reflection of what happens each month ... yielding about 210,000 annual visits.

Leaving Nordstrom

With mixed feelings, I have decided to leave my job as Vice President of Database Marketing at Nordstrom.

The positive side of this equation is that I am thrilled to be "running to something", not "running away from something". I will start my own business, appropriately called "MineThatData"!! My business will clearly explain the complex relationship between Customers, Advertising, Products, Brands and Channels to Multichannel CEOs and Executives.

My business will focus on the emerging field of Multichannel Forensics, a framework for understanding complex customer relationships coupled with a sophisticated forecasting tool used to understand the long-term sales, profit and growth potential of a multichannel business. Click here for a white paper on the topic of Multichannel Forensics.

I am so looking forward to starting this business!

This was not an easy decision to make. Nordstrom is a wonderful company to work for, consistently ranked as one of the top one-hundred companies to work for in the United States. I spent more than six years at Nordstrom, more time than at any previous company I worked at. That speaks volumes about the people I have worked with, and the environment at Nordstrom.

It will be particularly hard to leave the wonderful people who work in our Database Marketing department. My management team is among the most talented and experienced in the multichannel retailing industry.

• Few folks know more about multichannel circulation than Michael Einfalt. Not many folks manage teams that can explain the multichannel ROI of over 1,000 direct mail campaigns each year. Fewer people have linked clickstream behavior and retail purchasing to catalog response analytics. Almost nobody is a better manager of people than Michael. The latter is simply taken for granted. Few people truly see or appreciate his genius. Michael is too humble to brag about it.
  
• Few individuals are as innovative as Jennifer Thornton, our E-Mail Circulation and Online Analytics Manager. She has a rare combination of enthusiasm, energy, creativity, innovation, and an instinct to know whether something is 'right' or 'wrong', correct or incorrect. Jennifer has the potential to be a great leader in multichannel database marketing. I doubt anybody in our industry knows more about catalog circulation, e-mail campaign execution and analysis, and the use of social media in retailing. What a unique combination of skills. Jennifer is well-positioned for the future of our industry.
  
• Jay Long is our Director of Business Intelligence, the data mining, ad-hoc query and analysis arm of Nordstrom Database Marketing. No individual in the history of Nordstrom answered more questions about customer behavior (from a database) than Jay answered during the past six years. Jay completes more analyses per hour worked than any individual I have ever worked with. Couple that fact with his typical twelve hour day, and you have one of the most productive Business Intelligence individuals to ever work in multichannel retailing. Jay exhibits everything that is good about Nordstrom. Humble, honest, bursting with integrity and accountability, Jay has done more to make others look good at Nordstrom than any individual I have worked with. He sacrificed his career objectives to help others, always looking to support our Executive Team and our Decision Makers. Future leaders in Consumer Insights / Database Marketing will look great because of the foundation Jay built.

Linda Finn is our Chief Marketing Officer. She deserves huge kudos for providing an environment that allowed our team to do our job, without the micro-management and "do-it-my-way-or-the-highway" style exhibited by some leaders. She trusted our information, and trusted we were always trying to do what was best for Nordstrom. "LT", as she is known by, will go in the books as one of the better bosses I've worked for.

A few years ago, Jim Bromley was leaving his post as the leader of the online and catalog division of Nordstrom. Mr. Bromley asked me what I wanted to accomplish during the remainder of my time at Nordstrom. I told him that I "wanted to see how the story ended".

My first assignment at Nordstrom was to help Nordstrom Direct (catalog + online channel) become a profitable arm of a multichannel retailer. Several management teams later, the mission has been accomplished!

My second assignment at Nordstrom was to integrate separate teams that analyzed customer information and managed circulation for different divisions. Our employees did an exceptional job of building a team that looked at "one customer", across all Nordstrom channels.

My final assignment at Nordstrom was to use customer information to help our transition into multichannel retailing. I got to see what happens, politically, professionally, and financially, when a traditional catalog program ceases to exist. I learned what happened to the online business and the retail business when a catalog program is shut down. I observed what happened to customer acquisition when all traditional catalog acquisition activities end (list management, list brokerage, compiled lists, etc.). What a valuable learning experience!

Truthfully, I learned more about the impact of multichannel advertising during the past two years than I learned in my first seventeen years as a professional. That kind of experience can only happen when you essentially shut down one of your channels.

Having survived these assignments, I got to see how the story ended. I am ready to tackle my next endeavor. Thank you to everybody at Nordstrom, for making the past six years so educational, so memorable, so intellectually stimulating. I highly recommend Nordstrom as an employer, and think the world of my team at Nordstrom.