Integrating systems is a favourite past time of many organisations. In our opinion this is often overdone with integrations between everything that has an API. This can lead to stagnation of systems control that quickly degenerates into groups of them being ‘black boxed’ as they are too important to destabilise. This style of approach can lead to the kind of issues that Finance and Risk departments experience with different interpretations of the ‘same’ data, inevitably giving rise to issues with regulation such as GDPR, FRTB, and BCBS239 to name but a few.
It is imperative to first identify which systems master the data and which ones contain copies, augmentations, amalgamations and incorrect versions – then we must ensure that these masters are governed correctly. This will lead to correctly integrated, understood and governed systems that will drive efficient development of systems and processes, as well as ability to adhere to regulation and drive down operational costs. The aim is to reduced system numbers and eventually remove all end user systems which are often in excel or access databases and live on people’s hard drives.
Beyond the correct identification of systems, and the technical challenge of integration, data governance – driven by the unified data strategy – must be adhered to in order to determine what is integrated.