Smith & Williamson has a small development team within IT, which provides data to the business, from their legacy UNIX system using COBOL. It has some stand-alone windows-based applications but they largely access data from the legacy HP systems. The process for getting data to the business could be slow. A typical process would be a user request for data which would lead to an instruction for IT to write a COBOL programme. This programme then needed to be tested before data could be delivered to the user. As such, time to delivery could be several weeks. As it was difficult to use business intelligence (BI) tools over the legacy systems, the business had created sophisticated Excel spread sheets and Access databases to analyse data, many performed similar functions and tried to find answers to the same questions. The challenge for IT was to provide a single version of data that could be trusted and used for BI by the business.
The WhereScape RED data warehouse application was chosen to automate the entire data warehouse lifecycle, from design and implementation to operation making it agile, fast and cost-effective to deliver. David Rutherford, Head of Application and Data Architecture at Smith & Williamson had first-hand experience of using WhereScape in his previous role at a global risk management company and had been impressed by it.
Through interactive development, WhereScape delivered in three weeks what we estimate would have taken several months using a traditional approach with hand crafted code. As well as reducing the time it took to formulate reports, the data warehouse improved data lineage, traceability and documentation. The pilot also helped to identify areas for data quality improvement, which meant these could be addressed by the business.
It was key that the entire process was successful as it is a major component of both our systems and data architecture strategy, I have been involved in a number of data warehouse and ETL projects in the past and this was by far the quickest and most straight forward I have encountered and so I am delighted we opted to work with WhereScape.