From 1989 to 1994, the University of Chicago Booth School of Business and the now defunct Dominick’s Finer Foods entered into a partnership for store-level research into shelf management and pricing. Together, they conducted randomized experiments in more than 25 different categories throughout all stores in this 100-store chain. As a by-product of this research cooperation, approximately nine years of store-level data on the sales of more than 3,500 UPCs are available through this resource.
These data are for academic research purposes only. Users must acknowledge in their working papers and/or publications the Kilts Center for Marketing at the University of Chicago Booth School of Business.
The Dominick’s dataset covers store-level scanner data collected at Dominick’s Finer Foods over a period of more than seven years. The dataset contains two types of files: category-specific files and general files. The general files contain information pertaining to all the categories in the project.
The DFF dataset files, in zipped PC SAS format, are available for download. Please see the links below for more information on download locations. For more information, refer to the data manual.
Please refer to data manual for customer count file details, including variable names, descriptions, types, and lengths.
The customer count file includes information about in-store traffic. The data is store specific and on a daily basis. The customer count data refers to the number of customers visiting the store and purchasing something. Also in the customer count file is a total dollar sales and total coupons redeemed figure, by DFF defined department. These figures are compiled daily from the register/scanner receipts.
Download the ccount file in the following formats:
for SAS V6 (ccount(sas6).zip)
for SAS V7 or higher ccount(sas7).zip)
for Stata (ccount(stata).zip)
The demographics file consists of store-specific demographic data. The data originally comes from US government (1990) census data for the Chicago metropolitan area. Market Metrics processed this data to generate demographic profiles for each of the DFF stores. The table below gives the descriptions for all the files in the demographics database.
Please refer to data manual for customer count file details, including variable names and descriptions.
Download the demographics file in the following formats:
for SAS V6 (demo(sas6).zip)
for SAS V7 or higher (demo(sas7).zip)
for Stata (demo(stata).zip)
There are two files for each category studied in the course of the project. Each category is referenced by a three-letter acronym detailed in the manual, which you can download here.
The UPC files contain one record for each UPC in a category (xxx stands for the category acronym). They contain information about product name, size, commodity code, etc. The files are sorted by UPC.
The movement files contain weekly sales data for each UPC in each store for over five years. The variables included in these files comprise: price, unit sold, profit margin, deal code, etc. The files are sorted by UPC, store, week.
Category files can be downloaded by UPC or movement. Please refer to data manual for category details, UPC and movement data descriptions, file content, data organization, store numbers, dates for each week of data recorded, and other remarks.
The UPC files contain a description of each UPC in a category. The files are named upcxxx, where xxx is the three-letter acronym for the category. *Note: This is historical data and that the products referred to in the UPC files are not for sale.
The movement files contain sales information at the store level for each upc in a category. The information is stored on a weekly basis. The files are named wxxx where xxx is the three letter acronym for the category.
Use the UPC and Movement links below to download the category-specific datasets. Remember that the UPC files are named upcxxx and the movement files named wxxx where the xxx refers to the three-letter acronym used for file or variable naming purposes.
We have also provided these files in CSV format to make them more useful to researchers. Click here for the code used to convert SAS files to CSV. Note that the CSV files include the 'truncated' PRICE and PROFIT variables as well as the full precision in hexadecimal notation. Hence, the CSV files contain two new variables, PRICE_HEX and PROFIT_HEX, which if used give identical results as the original SAS files. All results based on the SAS files will be 100% replicable using the CSV files, i.e. there is no loss of information between conversion. Jens Mehrhoff completed this conversion and he suggests using the 'truncated' versions.
Note that there is no CSV file available for refrigerated juices.