Disney discovers a whole new world of data warehouse best practices

The director of data warehousing and analytics at Walt Disney Parks and Resorts shares some important BI and data warehouse best practices.

Childhood-dream-maker Walt Disney Parks and Resorts is gaining "actionable intelligence" from thousands of merchandising operations, thanks to an internal project that stressed the importance of adhering to data warehouse best practices.

The project, which was partially completed last year, aims to centralize the management of Disney's global merchandise operations onto a single enterprise resource planning (ERP) system affectionately known as SIMBA, or single integrated merchandise business application.

"We can have a lot of fun with our project names," said Juan Gorricho, the director of data warehousing and analytics at Walt Disney Parks and Resorts. Gorricho spoke about the SIMBA project at the recent Enterprise Architecture and Data Warehousing 2011 conference in New York.

Disney takes in nearly one billion dollars in merchandise revenue each year, Gorricho said, and building an ERP system that could handle all that information would be a challenge. That's why Disney initially focused the project on its U.S.-based theme and water parks. But eventually the system will encompass parks in Tokyo, Hong Kong, Paris and a new park currently being built in Shanghai.

The newly centralized ERP system was designed to handle merchandise management, inventory management and related business processes. But Disney also wanted to leverage the system for financial and business intelligence (BI) reporting and business analytics – and that meant building a new data warehouse.

"The purpose of the data warehouse from the angle of this project was to support analytical reporting for the most part [and key business processes]," Gorricho said. "There is a lot of value to capture from this data."

The newly centralized ERP, data warehouse and analytics system is helping Disney do a better job of managing inventory, analyzing sales numbers and predicting demand for merchandise in specific areas. Some of the products Disney used for the project include SAS Institute analytics software and Teradata data warehouse technology.

Gorricho said the project represented several "firsts" for Disney: Namely, it was the first time a merchandise enterprise data warehouse was used for financial reporting. As a result, the Disney team gained an appreciation for some specific data warehouse best practices along the way, and Gorricho shared them with the conference crowd.

Approach the data warehouse project from the business side of things. In the past, Disney's IT department was largely responsible for producing BI reports and analysis for business users. But Disney wanted to change that and make the business users largely responsible for creating their own reports. To accomplish this, the company decided to make some structural changes designed to emphasize the technology unit's new focus on business needs.

Before beginning the SIMBA project, Disney changed the name of its IT department to the "Global Business Technology Strategy" unit – a move that set the stage for IT to approach the data warehouse and analytics project from a business perspective, Gorricho said.

When modeling the data warehouse, the BI and analytics reporting capabilities, and any related business processes, Gorricho said it's important for technology workers to keep several questions in mind, including: What is the business trying to accomplish? How do business users intend to use the data in the data warehouse? Is there anything that can be done from a data warehouse perspective to make things easier on the business?

"Before, the business was just this remote client and we were filling [orders] for them," Gorricho said. "They now are seen as partners."

Use the most important data only. When Disney's technology workers first began setting up the data warehouse, the initial strategy was to bring in all the merchandise data from all over the country, including 450 database tables and "I don't even know how many terabytes of data," Gorricho said.

It soon became apparent that such a vast amount of information would be too difficult to manage, Gorricho explained, so Disney employed a new strategy: Concentrate on the merchandise-related data that will help business people accomplish their goals. In Disney's case, that meant bringing the information necessary to complete financial, forecasting, inventory and other business reports.

"It just doesn't make sense to bring the whole thing into the data warehouse," he said.

Establish checks and balances. It's important for business users to fully trust the information stored in a data warehouse. That's why Disney created a sophisticated balance and audit control system to ensure the data in the data warehouse matched the data in the financial system of record.

Discrepancies continue to pop up every once in a while and it's important to be prepared to quickly fix the problem, Gorricho said. The silver lining is that such errors usually provide an opportunity for business and IT to work together to create proper and long-lasting solutions.

Keep BI and analytics reporting simple. Organizations that want rank-and-file business users to make use of self-service BI capabilities had better make sure that they're easy to use. Gorricho said that means making sure the reports are targeted at business users’ needs, easy to navigate and, most importantly, easy to produce.

Technology workers need to have a solid understanding of which business users own specific dashboards and reports and they should customize the system with long-term business needs and "maintainability" in mind.

"We built simple solutions that the business could easily manage afterwards," he said.

Dig deeper on Enterprise data architecture best practices

Pro+

Features

Enjoy the benefits of Pro+ membership, learn more and join.

0 comments

Oldest 

Forgot Password?

No problem! Submit your e-mail address below. We'll send you an email containing your password.

Your password has been sent to:

-ADS BY GOOGLE

SearchBusinessAnalytics

SearchAWS

SearchContentManagement

SearchOracle

SearchSAP

SearchSOA

SearchSQLServer

Close