Data dilemma:
Banking's D-Day landings

Consolidating data onto a single platform sounds straightforward, says BRIAN BRODIE. But, “like the D-day landings, the devil lies in the detail”. (And he should know: he’s completed 66 data migrations and transferred 18 million-plus individual items over the past two years)

When William and Mary affixed their royal seal to the first charter of the Bank of England in 1694, little did they know that they were about to create one of the biggest data management challenges known to mankind.

Today, there are billions of bank accounts generating trillions of transactions across the world. If that sounds like a management challenge, consider the fact that hundreds of different banking systems have been developed to process all those accounts and transactions. The banking industry is awash with a multitude of systems, which makes managing data in a cost-effective manner difficult.

Even within an individual bank, there are often dozens of different systems managing myriad products and accounts. Last November, Chartered Banker reported that when HSBC instigated a project to migrate all of its legacy systems onto a single ‘gold suite’ technology platform, it involved replacing 55 separate core banking systems, 24 credit card systems, 41 internet banking systems and 40 desktop standards as part of the process. HSBC is not exceptional; in fact it’s very typical of most large financial institutions, which are fighting a constant battle to ensure legacy computer systems remain fit for purpose.

Consolidating data onto a single platform sounds as if it should be a straightforward exercise, but in reality it is anything but. Different systems use different programming languages and when some were built, modern products, such as offset mortgages, were not even envisaged. It isn’t, therefore, simply an exercise in lifting data from one platform and depositing it on to another.

Keeping on top of legacy technology is not the only factor exercising minds in modern financial institutions. Banks operate under a demanding regulatory regime, in which they are expected to implement new rules and regulations quickly, which means they need systems that can accommodate changes easily.

The last few years have also seen an unprecedented number of mergers and acquisitions, some of which have been hastily executed in very short timeframes. These have left banks with a disparate range of mortgage and loan books, with some being core to their business strategy and others being peripheral. Spending time, money and effort integrating peripheral loan books onto a core system can take valuable resources away from mainstream activities.

The large number of securitisation and whole loan sale transactions, which took place prior to the credit crunch, also resulted in loans being transferred to and from a range of platforms. Increasingly, this market is returning and many banks will want to securitise again, but their systems will be unable to handle the extra demands this places on them. The cost of developing a securitisation capability is one thing, but the real issue is how long it will take. Every month that passes has serious balance sheet implications.

Perhaps most importantly, banks now operate in an era in which having a compelling service proposition is of paramount importance. Financial institutions are re-engineering processes and procedures to ensure that their customers’ experience, from first contact and then throughout the lifetime of their accounts, exceeds their expectations. This places huge demands on some technology systems, which may constrain a bank’s ability to provide best-in-class service.

All of these factors create a strong case for banks to develop a single operating platform and then migrate customer data from existing systems onto that platform. Like the D-day landings, the concept is easy enough to understand but the devil lies in the detail, which is why many financial organisations decide to outsource this task to specialist companies. Successful migrations are underpinned by experience and learning from some hard knocks in the past.

Although the initial instinct of IT directors may be to manage data migration projects inhouse, the resources required to do so, in terms of both budgets and skilled staff, often exceed those available within their IT departments. What’s more, IT directors understandably want to retain valuable resources to manage other critical projects which are often time-sensitive and core to their business strategy. The availability of IT resource is always an issue in financial institutions.

Data migration almost always involves three key phases: phase one involves mapping existing processes to understand where differences and gaps exist on the multiple platforms in use. We find that it’s not uncommon for there to be a dearth of documented information about the systems in use, especially when loan books have been acquired.

After the data mapping exercise has been completed and documented, the second phase is to analyse the way in which an organisation intends to deliver services to its clients, so that appropriate administrative processes can be accommodated on the new platform. This is the phase when data migration exercises inevitably involve organisational and operational design input, as it’s an ideal time to review and revise existing procedures.

The third and final phase involves migrating data from old to new systems. If phases one and two have been completed correctly, then phase three should be trouble-free!

As banks and building societies face-up to the challenge of migrating data from legacy systems, it is inevitable that they start to question whether moving some customer data onto their core platform represents a good return on investment. For some, the conclusion they arrive at is that keeping multiple systems under their own roof and within their own control is no longer a practical option, because it’s too expensive and requires retention of a broader IT skill set.

Once a decision has been taken not to move some customers onto a core platform, then organisations have two options: firstly, they can outsource their back office or, alternatively, get someone else to run their IT, which enables them to retain control of all those processes which involve client interaction, but which removes the cost of maintaining an IT infrastructure. This latter option is becoming increasingly popular, as companies recognise that their core skills lie in product marketing and client relationship management, rather than in maintaining ageing IT systems.

Data is a core component on which every bank depends; it’s a fundamental part of their DNA. Managing data and successfully migrating it between platforms is, arguably, the greatest challenge currently facing any financial institution.

Brian Brodie, a Fellow of the Chartered Banker Institute, is Chief Executive Officer of HML, a leading UK financial outsourcer with 30 UK clients and more than £40bn of assets under management.

Back to Special Report contents
Back to Magazine contents