Royal Mail delivers enterprise master data management
The Royal Mail has delivered a master data management programme stemming from a new electronic data warehouse and BI initiative. Project team leaders claim a people-plus-technology synthesis.
The head of management information at the Royal Mail and his Accenture consultancy colleague see their recent enterprise master data management (MDM) programme as an example of how to marry people and technology to the detriment of neither and the benefit of both.
This week, at IRM UK’s conference, Data Management, Information Quality and Data Warehouse & Business Intelligence, they outlined how the Royal Mail has achieved an MDM implementation through a business intelligence project that was built around a new enterprise data warehouse.
Phil May, at Royal Mail, and Duncan Slater, manager at Accenture Information Management Services, gave a talk at the conference titled “Leveraging an enterprise data warehouse project to deliver enterprise MDM.” They spoke to SearchDataManagementUK ahead of the conference.
In June last year, Royal Mail started a supply chain enterprise data warehouse project with Accenture as the delivery partner. The first release of the reporting system was launched to 150 analysts and 1,000 business users in April. In the same month, the master data management governance system was also released.
The organization’s data warehouse was aging, Slater and May said, in the interview. Its replacement fits with the Royal Mail’s overall technology transformation programme envisaged by former chief information officer Robin Dargue and carried forward by his successors Yvonne Ferguson and Catherine Doran.
The 2008 Hooper report on the Royal Mail, Modernise or decline also said that improvements could be made with management information systems.
“The strategy behind [the electronic data warehouse] was that we had a multitude of different reporting platforms, and the common thread from the users was that the numbers coming out of the different systems did not tend to match, so therefore the data was not trusted,” May said.
Accenture addressed the problem by providing a platform that has consolidated information housed in disparate data sources, Slater said. The process involved sorting the master data and getting business units to agree on the definitions of terms.
“There was a lack of consistency around the definitions of various terms within the organisation," Slater said. "So even if you put a lot of human effort into combining [information], you were not always adding together apples and apples because there was no common definition of what, as an example, a 'delivery office' was.”
Before the MDM project, the various reporting systems suggested the organization had between 1,300 to 2,600 delivery offices but once the team settled on a particular definition of the phrase, they were able to see the correct number.
Accentures’s Slater reported that the MDM was delivered using an agile methodology. It was also delivered in releases, each focused on a specific aspect in order to manage the master data in a holistic way. The first three releases covered data definitions, bringing all the data together in one place, and the data governance workflows. The last release release added the ability to connect with additional source systems as well as "intelligent matching" capabilities.
IBM’s technologies InfoSphere Data Stage, MDM Server, and Lombardi for workflow, have been used together with a Teradata data warehouse.
On the people side, May reported that the organisation is training two data stewards in each of 11 regions and another six nationally. A data steward serves as a point person who can enforce data governance policies and settle disputes when disagreements over definitions arise.
"MDM is something that people think they understand completely, but there are different levels of implementation. You can do it in a very technology-centric way and you can do it in a people-centric way, but to be successful you need to bring together those things and strike the right balance," Slater said. “For example, you can do a lot of things manually, but then you lose some of the functionality that you bring into play with the automation that reveals where bottlenecks are.”
For May, the new architecture also means that the business people at Royal Mail can make workflow changes without recourse to the IT team in India. There are now clearer demarcation lines between business and IT, with more self-service capability available to the former.