article banner

How is big data modernising risk control management?

Jamie Crossman-Smith Jamie Crossman-Smith

'Big data' has changed the face of business. Chitvan Jindal from our data and audit innovation team looks at how it’s driving new trends in risk control management.

If 2020 has taught us anything, it’s the need to be responsive. Whether it’s a local lockdown, a national lockdown or final preparations for Brexit; most businesses have felt the pressure to act quickly.

Those decisions, however, must be based on accurate risk data, available at the right time. You can make the most of your data with easy-to-use robotic process automation (RPA) and artificial intelligence (AI) tools. These are generally no-code applications, which can be used by people without programming knowledge, and will automate your controls and apply continuous monitoring.

These are not new ideas. Automated controls and continuous monitoring have been around for a while, but they’re becoming more accessible than ever before.

In fact, if you have the Microsoft 365 suite, you already have the tools you need to get going, and more importantly, have already paid for them. The standard package now includes Power BI and Power Automate.

Power BI can aggregate and visualise data in real time, and Power Automate is a lightweight RPA tool. These are valuable programs, but all too often they are underused, with people scratching their heads over what on earth they’re for.

We take a closer look at how you can use these tools, and others, to automate and monitor your controls.

Getting started with big data

Business processes develop over time, so most organisations use a combination of emails, spreadsheets, and specialist software to track repeatable tasks. That makes it difficult to apply a standardised, repeatable control to each activity and to monitor the effectiveness of that control.

The sheer volume of operational activities can magnify the problem, so gaining a snapshot of business risk can be challenging if you need to make a decision quickly.

The trick is to:

  • standardise these activities to give a repeatable and structured output to support control automation using RPA programs like Power Automate
  • aggregate and analyse these outputs to flag any anomalies and monitor controls with visual dashboards like Power BI

This is a big win for your first- and second-line functions. Moving from manual to automated controls gives your team more time to focus on the wish-list items that help your business innovate and grow. It also takes the pain out of control monitoring.

Historically, your team would review every instance of a control within a data sample to check its effectiveness. A continuous control dashboard supports substantive testing, making it easier to find anomalies, and address them before they escalate.

While these tools do need regular oversight and review, they’re less resource-heavy and you only need to develop the program once, with occasional updates as the business changes.

Building the big data foundation

Creating your dashboard is relatively straightforward. The real challenge is getting your data in a useable format.

First off, there’s the job of structuring your data. You can express structured data through pre-defined outputs, which are machine readable to build control rules. This is easier said than done, and you may need to re-jig some of your processes.

Mapping your operations will identify the source of data and any processes that need updating. Having enough underlying data is also crucial, including good examples of anomalies to help inform the model.

There's also the issue of skillsets. Knowing where to start can be daunting and it’s not immediately obvious how to draw data from different sources or combine that information.

Additional training can give you the tools to develop an in-house solution, but you may find an outsourced control-as-a-service (CaaS) more convenient. CaaS generally comes as a ready-made program, but there are usually options to tailor it to meet your needs.

Developing use cases

Building internal skillsets can also help identify use cases and business areas where technology can add value. This relies on having a good working knowledge of the techniques used and how they can interact with your operational processes.

To give you an idea of some typical use cases for automating controls and monitoring dashboards, some examples include:

SOX monitoring

You can aggregate data from bank transactions, invoice approvals and your general ledger for at a glance reconciliation and tracking.

Cyber security

You can monitor threat data from all external domains and make sure all certificates are up to date.

Financial accounts

You can monitor accounts payable and receivable, with metrics on service level agreements, major debt holders and aged debt amongst others.

Our data analytics team can work with you at every step of the way and help you identify use cases, update your operational processes and structure your data.

For more information on our CaaS and control automation options and how we can help you, contact Jamie Crossman-Smith.

Blockchain and AI: the future of anti-money laundering? Find out more

Data analytics and COVID-19: making informed decisions

Find out more

Do you have an audit framework in place for AI?

Find out more