Ž . Decision Support Systems 29 2000 11–20 www.elsevier.comrlocaterdsw Management issues in data warehousing: insights from the Housing and Development Board James Ang, Thompson S.H. Teo ) Department of Decision Sciences, Faculty of Business Administration, National UniÕersity of Singapore, 10 Kent Ridge Crescent, Singapore 119260, Singapore Accepted 6 December 1999 Abstract Data warehousing has emerged as one of the most powerful tools in delivering information to users. In this paper, we Ž . examine data warehousing at the Housing and Development Board HDB , which is responsible for providing affordable, high-quality public housing to Singapore citizens. The HDB embarked on building a data warehouse because access to the diverse and large amount of data in its operational systems, was becoming increasingly cumbersome and time consuming. By building a data warehouse, the HDB aims to facilitate users’ access to corporate information for planning and decision making. The experiences and lessons learned from building a data warehouse at HDB are discussed. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Data warehouse; Lessons; Singapore; HDB 1. Introduction Over the years, firms have accumulated a large volume of data that potentially contain valuable in- formation about their business. These data, which are stored in operational databases, are not easily acces- sible to decision makers due to most firms’ existing Ž . information technology IT infrastructure. Further, ) Corresponding author. Tel.: q 65-874-3036; fax: q 65-779- 2621. Ž . E-mail addresses: fbaangsk@nus.edu.sg J. Ang , Ž . fbateosh@nus.edu.sg T.S.H. Teo . these operational databases are likely to be geograph- ically or logically dispersed, thereby making it diffi- cult and inconvenient to access. Ž . Traditionally, the information systems IS depart- ments are the sole interface to firms’ data stored in computer systems. Executives from various depart- ments rely on the IS department to satisfy their needs for information that is necessary for decision mak- ing. The turnaround time is usually long and infor- mation is often not delivered on time to users, which in turn reduces the value of the information. Further, reports produced by the IS departments usually pro- Ž vide a one-dimensional rather than multidimen- . sional view of the data. For example, sales can be broken up into geographical districts, regions and 0167-9236r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. Ž . PII: S0167-9236 99 00085-8