Data Sets

Data sets are made available to approved academics for classroom use, dissertations and/or other research and are free of charge to members of the Marketing EDGE Professors’ Academy. Data set usage rules may vary. Free online access to data sets is available for Members of the Marketing EDGE Professors’ Academy. Cost to non-members is $25 per data file. The data in the library is provided “as is.”  We cannot resolve discrepancies in the data nor can we provide information beyond what is included.  A Non Disclosure Agreement is required in order for us to release the data sets.

Data Sets (1-6) contain customer buying history for different database marketing businesses and two ZIP Code data files. Each data set contains actual customer behavior as tracked by the business organization for about 100,000 customers. Data is available in ASCII, SAS, or SPSS.
Data Set 1 is from a non-profit organization that uses direct mail to solicit additional contributions from past donors.
Data Set 2 is a business with multiple divisions, each mailing different catalogs to a unified customer base.
Data Set 3 is a long-time specialty catalog company that mails both full-line and seasonal catalogs to its customer base and often re-mails the same catalog to its best customers.
Data Set 4 is an upscale gift business that mails general and specialized catalogs to its customer base several times each year.
Data Set 5 contains data relating to 42,765 ZIP Codes summarized from credit information available for a large sample of households at Experian; one record per ZIP Code.
Data Set 6 contains information relating to 34,297 ZIP Codes primarily based on 1990 Census or 1995 Census Update with some proprietary additional fields available to the data provider; one record per ZIP Code.
Data Set 7 – A sample of this Data Set was used in Marketing EDGE’s 2008 Customer Lifetime Value Competition. Data is from a leading US Charity and contains the donation and solicitation history for over 1 million donors. The donation history spans 14 years (1993-2006), including the donor ID, donation date, date of the first donation, amount of donation, the 5-digit Zip Code of the donor and solicitation ID. The solicitation history spans 15 years (1992-2006), including donor ID, solicitation ID, and the solicitation date. The data set also contains the costs of individual solicitations.
Data Set 8 is from a catalog company that mailed seasonally to existing customers, customers of subsidiary/affiliated companies, and customers reached via web advertising. The catalog was promoted through both direct mail and email. The Data Set contains twelve years of data, through April 30, 2009. Information includes catalog orders, order source, quantity of items purchased, returns, payment information, and the Zip Code of the purchaser. There is one record per order, with multiple orders per household. Orders with the same household are indicated with matching Household-ID numbers (one number per unique household.) The file contains 14, 448 order records from 10,000 unique households.
Data Set 9 is for classroom teaching and is from a multichannel company with sales of several hundred million dollars per year. This nationally known company has a network of retail stores, a well established traditional catalog channel and a website. The majority of its sales are food products purchased as gifts during the Christmas season. This Data Set includes over 100,000 customer records and over 3.3 million marketing contact records.
Data Set 10 is from a leading US provider of community-based preventive health screenings of over one million people each year at over 20,000 events nationwide. They are working to rapidly grow their customer base by promoting the dates and location of their screenings through targeted promotional mail advertising.
Data Set 11 is a dataset for classroom teaching only. This dataset is from a cooperative database company and compiles household purchasing data from hundreds of companies. It contains nearly 2.5 million households from the overall 13+ million-record database and covers the period from January 2005 through December 2007, including extracted order and line item records. Reference tables with product area, major category descriptions, and records with the ZIP Code of each household are provided. Based on the extracted data sample, students could be asked to address marketing issues such as:
  • Developing and comparing predictive models for house file marketing selections.
  • Building new customer acquisition selection systems. Students could be asked to focus on a particular participant company and find ways to use the database information to predict which households will become new customers.
  • Companies examine market trends by merchandise category to develop merchandising strategies. Students could use the data set to understand trends affecting businesses.
  • Identifying affinities among participating companies; that is, which companies appeal to many of the same households and are, therefore, possible competitors.
  • Tracking and comparing customer sales value over time, switching channel loyalty over time, and/or seasonality issues.
Data Set 12 is from a leading subscription-based TV company with a sample of 100,000 of its customers (available in SAS format only). The data set includes basic customer information - channels and networks subscribed to, associated demographic information (such as children and children’s ages in household), education, specific income levels e.g., ‘blue collar household’, computer in household, languages spoken, market value of car, etc. It also includes behavioral information such as interest in automotive-related activities and magazines, gardening, photography and more, etc. – over 60 categories in total. An Excel spreadsheet contains tabs featuring 1) a data dictionary, 2) variable distributions, and 3) data contents.
Data Set 13 - This data comes from an organization with a health related mission.  With more than 4.3 million transactions from 2007-2010, the data set contains two fields for each transaction, which indicate the appeal that the contribution pertains to.  An Excel Spreadsheet specifying the variable fields (comprised of Account, Gift Date, Gift Amount, Appeal Description, and Appeal Type) is included.
Data Set 14 - Spanning from 2007-2010, this data consists of more than 1.3 million transactions from a children’s welfare organization.  Each record indicates the appeal the donor replied to, as well as the date of the donor’s first contribution.  Comprised of two files, the first contains a list of the field, the second contains the data.
Data Set 15 - A specialty multi-channel catalog company with a web presence mails a high volume of catalogs annually as its primary marketing vehicle.  100,000 random US customers make up the records, which include all orders, dollars, items, order lines and returns from December 16, 2004-September 17, 2012, as well as all catalog contact history during this period.  An included Word document suggests potential data uses, and includes field descriptions.

Marketing EDGE welcomes your feedback. Please help us to improve the content of the above data sets to better support your curricula and/or research by contacting us at

Access Data Sets

Professors' Academy...Learn More