When people start looking for solutions to financial crime, regulation and compliance there is a natural tendency to start with new customers, known as the ‘front book’. 

These new customers already expect to have to go through some onboarding processes, and are typically coming in at a manageable volume. They seem to offer a natural starting point.

But the uncomfortable truth is that, for established financial institutions, the vast majority of financial crime and fraud originates from pre-existing customers. It is the ‘back book’ that hides the biggest problems.

Why is this the case? For a start, there are just many, many more customers in the back book than front book. If a firm grows its customer base at say 5% a year there will be around 20 times as many existing customers in the back book. Given this difference in volumes it’s not difficult to understand where the majority of financial crime will take place.

Also back book customers will typically have been brought on board at a time when the relevant regulations and checks where less robust. Regulators haven’t missed this uncomfortable truth, and there is an increasing pressure for institutions to address this area.

Dealing with the back book entails a huge amount of work. Being able to analyse quality of its base data is an essential way to make achievable what could otherwise be a potentially overwhelming task. Given the large volume and potential for multiple, often long running, issues involved this task is not to be taken lightly. The key is to understand where and how to start.

For those expecting a discussion of complex Bayesian statistical techniques on big data supersized specially constructed data stores at this point, I am afraid I will have to disappoint you. In my experience of running these types of programmes, the way to address the back book is to start with simple, but highly relevant business driven tests on a set of key data elements. These tests will likely include elements for Anti Money Laundering and Sanctions, unique identification and potential red flags driving levels of due diligence. By running these tests across the whole back book you will get insights into not only its current quality, but also where to start improving it.

The exact plan to address these issues will vary from firm to firm, not least due to availability or not of external data for comparison and an ability to prioritise subsets for remediation, but a key mandatory step is to initially look across the whole back book to gain a fully informed view. Unfortunately, just because remediation of back book data is complex, difficult, time consuming and has a tendency to lead to requests to some customers for information you should already have does not mean it can be ignored.

If banks and other financial institutions have not realised it already, they are being forced by regulators to discover that an untamed back book can bite.