Royal Bank of Scotland (RBS) Group may have faced a substantial fine this month from the UK Financial Services Authority (FSA) with regards to failures related to its anti-money laundering (AML) checking process, but a lot of work has been going on in the background at the firm in order to get its customer database into shape, according to its data management team. Mark Davies, head of reference data at RBS, explains to Reference Data Review that the project that has been taking up the most amount of time this year has been focused on decommissioning the legal entity database acquired during the ABN Amro buyout and moving this data onto its centralised RBS system.
Davies explains that the firm has been engaged in decommissioning the ABN Amro system, which has around 70 or 80 different systems hanging off it, and moving the relevant data onto RBS’ existing legal entity database. “It has been quite a major job in terms of system feeds and testing, including making sure the data is set up correctly and adding in new fields and attributes, as well as getting rid of duplicate records so that there is one version of each company. That has taken a lot of budget and time over the last couple of years,” he says.
The RBS legal entity database currently spans the entire wholesale business and every legal entity in the RBS Group is tracked on the system. The firm opted to go in this direction when it began working on meeting its Basel II requirements: “When we started to look at our Basel II requirements, we rolled out a global banking system to be able to manage all of our wholesale customers.”
Davies indicates that this is not the norm in the market, as most other companies have centralised data management systems at a divisional level. So, for example, the investment banking division of a firm will have a centralised entity data system and the corporate banking division of the same firm will have a separate system for this data. However, he notes that firms of a similar size and complexity are also looking to move in RBS’ direction, notably Barclays.
One of the first priorities of the most recent customer data integration project, however, was to get the legal entities set up on the recipient system side and this first began around two years ago. That process then merged into bringing the two databases into line with each other and keeping them in sync, explains Davies. “It has been a combination of data matching and technical changes with a lot of manual intervention and reviewing of that data,” he adds.
The firm opted for in-house development of this project rather than a vendor assisted solution because of the importance of the central legal entity database to RBS’ business. “From an IT perspective, I don’t know if we will move to a vendor solution in the future, but I suspect we will continue with an in-house solution,” explains Davies of the potential to roll out a vendor provided entity data solution.
The challenges involved in the process have been numerous but one of the most time consuming has been around understanding the differences between the two data models, according to Davies. This has included defining the differences between what constitutes a legal entity in each system and how things are defined within those systems with regards to data formats, standards and naming conventions. “One item that may be considered duplicate in one system may not be considered as such in another, for example,” he elaborates.
However, the largest obstacle for the team has been in taking into account the legal considerations surrounding the move of ABN Amro data, which was bought by a consortium, to RBS’ in-house platform. Davies explains that the team had to navigate a number of legal considerations around data sharing, as there were only a portion of customers in the ABN Amro system that could be considered as RBS customers.
“Right the way through the process we have had a lot of restrictions about what information can be shared and we had to wait for some of the legal demerger to happen. ABN Amro was only legally broken up in March this year, so a lot of the work before that had to be carefully controlled and it was the same for the other two consortium partners,” he says.
The RBS team has taken a database that started out at around 500,000 legal entities and it is now made up of the best part of 1.5 million, so Davies’ description of the project as a “huge job” is seemingly justified.
And the work doesn’t stop there. Davies notes that maintenance of the data will be required and the team has duly put in place a number of new maintenance feeds and data quality metrics in order to ensure this work proceeds smoothly. “We have been doing a lot of benchmarking activity to see what the data quality within the system is like. We have got a large team in Chennai, which is made up of the old team from ABN Amro and the new RBS team, that is working to review the data quality and deal with any compromises that had to be made, making sure that they are cleaned up after the event,” he adds.
RBS started from scratch to set the data quality metrics in place because there are new fields and a whole new set of customers to deal with, explains Davies. ABN Amro has brought RBS a bigger global footprint and the firm is now dealing with a more countries than before, which brings its own data requirements.” One of the challenges has been understanding that a lot of the information and inputs is now coming to us in local languages rather than English. We have checked our business processes to be able to provide round the clock support to be able to cater to local language submission,” he says.