No More Day Old Bread

UP UNTIL LAST YEAR, HAGGEN SUPERMARKET was organizing its business intelligence data in almost as many ways as it plans departments for its food. While consumers embrace choices in meat and produce, the chain's complex collection of spread sheets and other data on weekly performance took at least 24 hours to prepare. The end results were as limited in use as day-old bread.

The 32-store grocery retailer had to delve into scattered information across a host of homegrown and manual systems that covered planning, finance, loyalty and human resources functions. The absence of a data warehouse and business intelligence (BI) system made the integration, transformation and maintenance of quality data problematic.

"We retained historical data that resided in many formats, such as Excel spread sheets," says Harrison Lewis, CIO of Bellingham, Wash.-based Haggen. "The analysis was very summarized. It could take three days following a sales week to obtain this data. There were also limits as to how we could use the information."

Complex merchandise and category performance analysis was impossible to report because multiple levels of granularity (including department, category, sub-category and SKU levels) and several source systems needed to be connected. Special reports, like volume reports, could not be drilled into or filtered. The low margins of the food and grocery segment required merchandisers and category managers to constantly monitor and analyze the sales and margin performance of food categories. It also was a challenge to perform pre-promotion during promotion and post-promotion sales comparisons.

Haggen Supermarket, which operates under the Haggen Food & Pharmacy and TOP Food & Drug banners, realized data-driven reporting and analysis could allow its business managers to make timely, informed decisions. Better information also could help the organization expedite its growth plans.

In June 2007, Haggen Supermarket deployed Teradata's integrated data warehouse and Manthan Systems' ARC Enterprise Business Intelligence. The platform included an enterprise retail data model and BI functionality that covers store operations, category management, inventory and promotions. The system also has been extended to human resources, payroll and loss prevention analytics. It was deployed in all 32 Haggen stores across Washington and Oregon in just 16 weeks.

"Implementing a data warehouse is not a destination, it is a journey," says Lewis. "The retail specific capabilities of ARC and the high performance of Teradata have given our business managers a powerful platform for retail analysis that aid their category management, operations performance and inventory management decision making processes. The rapid deployment of this solution has helped us realize business value much faster than we would have with other generic options. In our highly competitive business, we have to respond to the market very quickly and cost effectively. This solution has made it possible to do that."

Today, the retailer configures complex reports on its inventory and sales throughout the day. The grocer also has the ability to compare private label performance to that of national brands. The average reporting time has gone from more 24 hours to less than five seconds. The data warehouse, which holds five years of
historical data, has been deployed on HP servers operating in a Windows environment.

The food and grocery variant of the ARC retail data model can recognize specific nuances within Haggen's business. The retailer also uses ARC to enable item-level reporting on merchandise performance. ARC gives the perishables team, sales team and category managers visibility into movement, sales and gross margins across the merchandise hierarchy. Before, this information was unavailable.

Quick generation of reports, such as item and price comparisons, enable department managers to make better micromarketing decisions. Store managers use data to obtain information about what is selling, inventory status, out-of-stocks and market basket analysis, says Lewis. Financial analysts and controllers can analyze costs by separate buying and selling merchandise hierarchies.

Following deployment of the Teradata system, Haggen implemented GoldenGate software for Real-Time Business Intelligence to improve the quality and timeliness of its data integration process. The retailer uses GoldenGate solutions for real-time data feeds. This facilitates downstream reporting and business intelligence. Plus, it significantly improves timeliness of data and reduces transmission errors. Every few minutes, data is delivered to the Teradata Enterprise Data Warehouse. Information, such as revenue and margin by store (including information pertaining to a particular product), is then readily available for business-wide reporting and analysis.

"With GoldenGate in place, we have not only seen a dramatic decrease in data latency and errors, we have obtained visibility throughout the sales day in record time. This allows us to make important decisions about the business that have a positive effect on that day's sales and operations," says Lewis.