Improving data to solve financial crime operational challenges

Take one example. I recently worked on a major offboarding programme, hired because the firm was overwhelmed by the amount of work involved. Dormant clients were tying up valuable resources; analysts were spending hours chasing information from clients who hadn’t transacted in years. 

By taking the time to analyse the available data to identify and offboard these clients before they landed on an analyst’s desk, we streamlined the process and freed up critical capacity. All because we saw the client’s issue as a business challenge that had a data lens, and solved that first. 

Most of the Ops leaders I speak to recognise that better data could help them with any number of internal challenges. But they’re time-poor, and they don’t have access to people with the right skills. Data people are great at data but rarely understand financial crime; financial crime teams often don’t understand how best to use or connect their data to support decision-making. Problems remain unsolved.

If you're nodding along to this story, take heart. You’re not alone. In my experience, organising client data is an entirely solvable problem that is solved almost nowhere.

All too often, leaders reluctantly accept the status quo, managing challenges as best they can. Problems fester until they become critical – or worse, result in a regulatory breach.

So how would I advise a firm stuck in this scenario? I ask Ops leaders to start small.

Take one challenge that can be solved using your data, and figure out how to approach it. Then you can build from there.

To start, you need to consider the issue that you’re facing and ask yourself what data points could help answer your question. Broadly, client data challenges fall into one of three types:

1. Identifying client characteristics

Ask: What data attribute, or set of attributes, are the hallmarks of this particular category of client?

To solve your immediate challenge, you may need to identify the data markers that tell you their risk score, product usage, geographic exposure, or any number of other client characteristics. By mapping and often combining these data attributes, you can reveal previously hidden insights into your client base, laying the groundwork for the next phase of work.

2. Mapping client operational data

Ask: What are the markers in the data that confirm, with confidence, that a certain task or operational procedure has been conducted?

To answer a question about the current status of an internal process, like the offboarding challenge I recently encountered, you will need to understand what data attributes demonstrate operational progress or completion. That could be an open case flag, or perhaps the presence of an account type associated with the client. These attributes aren’t all in one system. Be prepared to get creative in linking them – trade system IDs, to LEIs, to account IDs and so on. Gaining insight into this data provides a depth of understanding that many don’t have and provides a level of certainty that is often missing. Imagine knowing every day, with certainty, which clients had traded, or remediated, or completed legal document negotiation. How valuable would that be?

3. Uncovering efficiencies

Ask: How can I use my data to create the most efficient process?

Efficient processes stem from well-understood data. Once you’ve defined the attributes and markers you need, the next step is using them to design smarter workflows. This is where grouping, streamlining, and prioritising come into play.

A great example of this is identifying major client groups – say, multiple entities under a single parent organisation. Instead of treating each entity as an individual task, consolidating your approach might save time and reduce duplication. You could communicate with that client once, provide them with a more joined up service, or even spot unusual activity in one entity and immediately be able to consider if it impacts the other parties. Is your organisation able to provide you with this visibility today?

Why do I suggest starting by understanding which type of issue you’re dealing with? Because you know your business or function and can envisage what is possible. From there, you can engage with your data teams to make it a reality. It’s the first step to solving your data problem.