In recent years, the number of companies offering health supplements, nutritional products, and preservative-free skincare and cosmetics has increased, resulting in fierce competition. According to Masaki Ikemori, corporate systems group, information systems division, administration headquarters, FANCL Corporation, the company needed a system that enabled visibility of customer details.

“Customers were not differentiated into specific groups, and it dawned on us that we were not offering an appropriate response to each individual customer,” he said. “We could only see how many customers we had, not what type of customers they were. For instance, we could send out a news magazine to tens of thousands of our customers, but the only key performance indicators we determined were the number of customers who read the magazine or who made purchases within a specific time frame. We couldn’t see the attributes of the customers, such as who they were and what was important to them.  

“We realized we needed a customer-based approach, whereby each customer is treated as an individual and approached in a personalized manner, along with a system that made such an approach possible.”

FANCL Corporation also wanted to expand its customer and sales analysis capabilities.

“Our previous analysis system was not user-friendly,” said Takuto Watanabe, corporate systems group, information systems division, administration headquarters, FANCL Corporation. “To analyze data, such as the number of a particular lipstick sold, staff needed to be knowledgeable, not only about data infrastructure, such as where customer and product data is stored and how to access it, but also which analytical tool among several would be the most appropriate for a particular type of analysis. As the system was so complicated, only a few people could use it.”

Another issue was that the analysis system was divided up by sales channel with one analysis system for each channel, which made it impossible to see if a customer purchased from mail order only, or if they also shopped online and/or in-store.

According to Rio Endo, corporate systems group, information systems division, administration headquarters, FANCL Corporation, “There were cases where marketing campaigns aimed at mail order channels contributed to in-store sales, such as customers who purchased a three-month supply of a supplement online but their next three-months’ supply in-store, but we had no way of measuring the correlation. Mail order campaigns were measured simply by the results seen from mail-order sales. We needed a system that would enable us to carry out a comprehensive analysis of what influences a particular campaign could have on customers’ buying behaviors by channel, such as whether customers were buying a health supplement online that was being promoted through a mail order campaign.”
Finally, the analysis system also suffered from performance issues. “A single analysis, such as a demographic query regarding age or location, would usually take between 30 and 60 minutes,” said Ikemori. “Analysis involving large quantities of data, such as the extraction of specific demographic targets with a particular combination of gender, age, location, and purchasing traits, would often take up to a week, including manual preparations.”

To resolve these issues, FANCL Corporation reformed its information systems to create an environment that allowed for the rapid analysis of large quantities of data and would enable people not well versed in IT to carry out advanced analytical procedures. To make this possible, the company required a data warehouse where information on customers and their purchases could be stored, a machine to process the vast quantities of data stored in this data warehouse at high speed, and the construction of a plan-do-check-act (PDCA) cycle related to marketing campaigns.

FANCL Corporation selected Oracle Exadata Database Machine X2-2 to serve as the platform for the data warehouse; Oracle Business Intelligence Enterprise Edition to conduct analysis; and Siebel Marketing, to evaluate the results of marketing campaigns. The company started implementing these solutions in June 2010 and went live in May 2011.

Ensuring the usability of the business analysis system was critical to the reformation project.

“Once marketing staff and other front-line users were able to analyze product sales by customer and sales channel, they could set up hypotheses and test them as rapidly as possible, deepening their insight into the business,” said Watanabe. “In addition, by sharing this knowledge across the organization, we could improve efficiency.”

In the past, the Information Systems Division had to produce a wide range of reports for users in the sales and marketing departments. With Oracle Business Intelligence Enterprise Edition, users can now create ad hoc reports themselves, lifting the burden on the Information Systems Division by reducing the need to generate similar reports for different users and divisions. This also makes it possible to compile and analyze reports immediately if there is an issue or business opportunity in a particular product sector.

In addition, the company has reduced the number of reports by three-quarters—from 500 to 120—creating a standard set of reports and sharing information across the entire organization, which saves time and resources.

The ease of use of the Oracle system has led to an increase in the number of users undertaking analysis. “In the past, around 20 to 30 people in the marketing department were analyzing data, but because the Oracle system is so user-friendly, it is now being used by around 300 people, or approximately half of all employees,” said Watanabe.

FANCL Corporation often runs marketing campaigns where gifts or discounts are offered to customers who reach a certain number of points. These points are collected under a customer loyalty program, and also allocated to customers who spend a certain amount of money within a certain period.

However, it was difficult to tell with precision whether these campaigns were effective. The company wanted to see whether it was meeting key performance indicators for all its campaigns but couldn’t tell which campaign was generating mail order response rates or walk-in customers by just using the results from the loyalty program.

“One of the reasons we could not measure the results of marketing campaigns was because the standards of evaluation differed according to the sales channel and the marketer in charge,” explained Endo. “We would measure the success of a retail campaign by the number of customers visiting a store, but measure a mail order campaign by the number of responses we received. However hard we tried, it just wasn’t possible to evaluate return on investment by using the same standards to analyze and compare the results of various campaigns.”

Siebel Marketing enabled FANCL Corporation to resolve these issues. The system stores the same information for different marketing campaigns, such as response rates, expected ROI metrics, and how much that campaign contributed to revenue. The results of campaigns can then be analyzed by combining and cross-analyzing this information with transaction data, such as product order figures.

“We wanted to look at what campaigns had been adopted for specific products in the past and what sort of response we had from customers,” said Endo. “By visualizing this information for each customer, we can develop marketing campaigns that are better targeted to specific customers. Siebel Marketing enabled us to achieve this goal by providing a 360-degree view of customers, from what products they purchased, to which channels they used, and how much they spent.”

Siebel Marketing has also allowed FANCL Corporation to measure and evaluate marketing campaign results across multiple sales channels, using standard evaluation criteria. This has enabled the company to complete more marketing PDCA cycles and gain a better insight into how to make the next campaign more successful.

“By looking at customers across different sales channels, we can see which campaigns are proving to be effective,” said Endo. “By doing this repeatedly, we can select those marketing tactics that deliver the best results. This is the PDCA cycle we’ve now set in motion.”

The analysis system has already become indispensable for various departments. The company has linked the Oracle system to its customer relationship management and efforts are being made to improve the quality of marketing activities. According to Endo, “We didn’t have clear criteria to measure marketing campaigns in the past. The new system has allowed us to establish clear standards for accurately evaluating individual campaigns.”
Having created a platform for visualization, FANCL Corporation intends to embark on projects involving other business systems, such as supporting storefront activities. By bringing about further evolution in IT environments, the company hopes to enhance its competitiveness yet further.

When analyzing marketing activities and campaign results, sales order data over the past few years had to be collated with customer data. However, the master data on customers and sales transactions was as much as one terabyte. The previous system had difficulty extracting the relevant information quickly and efficiently, so that a simple query on how many customers had purchased a cleansing oil advertised in a marketing campaign, for example, could take up to an hour.

Oracle Exadata Database Machine significantly improved the performance of the data warehouse. According to Ikemori, it now takes less than half the time to process the same amount of data.