The Distribution Economics Institute of Japan adopts SAS® to handle diverse data analysis needs

DEIJ creates system enabling rapid development of advanced in-store marketing research analysis

The Demand Chain Development (DCD) joint research, operated by the Distribution Economics Institute of Japan (DEIJ), is attracting attention from businesses and government agencies as an organization that develops effective point-of-sale (POS) marketing methods. The DCD's main activities are joint research, which involves surveys and trials targeting consumers and retailers to obtain knowledge regarding distribution and marketing; and data analysis, which collects and accumulates POS and other data to provide data analysis infrastructure for member businesses.

Recently, in view of the growing trend toward disclosure of POS data by retail chains, manufacturers' needs regarding analysis have diversified. However, DEIJ's conventional data infrastructure is becoming unable to handle increasingly complex analysis needs. DEIJ started operating a new data analysis system based on SAS and has achieved major results, including a substantial reduction in operating costs as well as meeting a variety of data analysis needs.

Urgent analytics issue
The DEIJ is a distribution and marketing think tank authorized by the Ministry of Economy, Trade and Industry (METI), and was established in October 1966 as a public-interest research organization. Through collaboration by member businesses in the retail industries, the institute conducts a wide range of activities. These include in-store research through surveys, trials and analysis at supermarkets, convenience stores and drugstores; joint research by surveying changes in the environment surrounding the retail industry; and joint surveys and consulting under contract from businesses and government agencies.

Says Hiroyuki Kato, Senior Researcher: "Recently, there has been a growing tendency for retail chains to disclose their POS data, and collaboration between retailers and manufacturers is becoming vigorous. In line with this trend, there is an increase in outsourced and joint research so that manufacturers can suggest optimal counter layouts to retailers, thus diversifying the needs for data analysis. However, our previous data infrastructure was not very flexible or scalable, which made it difficult to handle increasingly complex analysis needs."

Also, FSP (frequent shoppers program) data linked to customer IDs must now be analyzed in addition to POS data, which shows what products sold, at which store, at what time and in what amount. This has greatly increased the amount of data accumulating in databases. Consequently, the problem of declining performance in analytical processing has become serious.

DEIJ did not have an analysis menu available in an easy-to-handle SAS programming language.

So DEIJ started a system development project. The aims of the project were to organize and integrate databases, improve data analysis performance and rationalize master data management, as well as enable a flexible response for diverse analysis needs.

In selecting the system, reducing system operation costs was an important requirement. With the previous data infrastructure, programming was done in the S language, which did not have a command system and was difficult for DEIJ to handle. Therefore, systems had to be ordered from a system developer in order to provide new analysis menus. Another goal was to eliminate the need - and cost - associated with outsourcing the development of new menus.

According to Takashi Teramoto, Senior Researcher: "Although we evaluated several packages in terms of their compatibility with various requirements, the only viable options turned out to be SAS Enterprise BI Server, SAS Data Integration Server and SAS® Enterprise Guide®. The main reason we chose SAS is that we considered it possible to introduce SAS smoothly, and because over half of our staff are familiar with the SAS programming language, there was no particular resistance or obstacle."

When the development phase started, DEIJ first organized and integrated its master data, including merchandise, customer, store, POS and FSP performance data. Next, data formats were integrated, and rules for updating and history management of master data were clarified. Among the analysis menus in the S language, frequently used menus, including revenue, ABC analysis, purchaser attribute and revenue trend analyses, were converted to the SAS programming language. As a result, users could now easily access and manage the menus. By specifying data items and analysis sequences using SAS data management capabilities, it is possible to do cross-analysis of performance data, such as POS and FSP, and master data regarding merchandise, customers and stores. DEIJ also created specifications enabling flexible execution of analysis menus in line with diverse usage scenarios.

Major reduction in operation costs
In the basic analysis menu in the new system, DEIJ included revenue ABC analysis regarding sales records, revenue trends analysis, purchaser attributes analysis, brand sales prices analysis, trial usage rate and recurrence rate analysis, purchasing switch analysis, and related purchasing analysis. In addition to the basic menu, it was possible to quickly reflect more advanced analysis methods, developed through DCD joint research, in the new data infrastructure. When discussing the decrease in operating costs, Kato explains: "It became possible for analysis methods developed by researchers at the institute, who were familiar with the SAS programming language, to be quickly incorporated into the developed menu and reflected in the system. As a result, we were able to substantially reduce system operation costs, including the expense of outsourcing, which was consistent with our aim in introducing the SAS system."

Also, the 40 members of DCD can access the new data analysis system via the Web browser, and by utilizing a variety of analysis menus, this helps to develop counterdemand and increase productivity. Members can upload POS and other analysis data to databases via the Internet, greatly reducing the previous need for manual data entry. The data management capabilities enable POS data collected from retail chains to be stored in integrated formats, making the analysis process much more efficient.

Teramoto adds, "We can automatically e-mail weekly and monthly reports to members who need them, or members can download reports. Many members say this has enabled them to develop timely product and marketing strategies in line with rapidly changing market trends. So the evaluation from members has been very positive."

Looking ahead, DEIJ aims to further improve analytical performance by organizing and integrating the data and providing templates that reflect analysis procedures. It also plans to utilize the advanced analysis procedures of SAS Enterprise BI Server to provide an analysis menu for relating customers' life stages and lifestyles to their purchasing behaviors.

The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.

Copyright © SAS Institute Inc. All Rights Reserved.