Reporting and Governance
Chris Mullan

Chris Mullan

Chris Mullan, senior vice-president product at Eigen Technologies, a platform that provides automated data extraction and processing solutions, suggests an upfront focus on fundamental weaknesses in existing data will lead to better outcomes when seeking to improve organisational efficiency and effectiveness via new data solutions.

The 14 years I spent working in the financial services and insurance sectors showcased the unbelievable amount of mis-spend organisations allocate to their digital transformation efforts. During that time, I watched enterprises pour millions of dollars into shiny, new “solutions” that often failed to deliver the anticipated and required results, and that almost always bypassed root issues hindering efficiency gains.

It’s easy to see how we got here. Throughout the past decade, many innovation teams across financial service institutions lacked constraints when thinking about the future. They dreamed of initiatives that would dominate headlines and craved the latest technologies they believed were capable of getting them there. More often than not, however, this meant decision-makers invested in technologies before solving deeply rooted issues. For example, data enrichment became a priority despite the fact that most financial institutions hadn’t quite figured out data extraction (and still haven’t).

It’s a twist on a classic idiom: putting the data enrichment cart before the data extraction horse. This miscalculation now poses severe problems for financial service companies looking to grow and modernise in the years ahead.

Scalable inefficiencies wreak havoc on profitability

Picture this: you work in wealth management advising more than 500 clients about their overall financial health and investment decisions. You review complex and disparate documents daily, from your clients’ insurance policies to brokerage statements. 

But if you make an error while manually inputting client data into your company’s data lake, your risk management team is now writing a report based on inaccurate information. Because of the financial service industry’s large-scale infrastructures, a seemingly minor issue can quickly snowball into a much larger problem. In this scenario, the root issue is erroneous manual data extraction and entry. The problem? Many of the shiny tools companies have invested in are better suited to addressing data needs and applications after this type of damage is already done. 

In an industry flooded with massive capital expenditure budgets, it’s time for decision-makers to rethink their technology investment strategies. The mindset around digital transformation must shift from a visionary approach to a problem-based one. There’s no room for frivolous investments. If your current technologies and processes prioritise data enrichment, it’s time to go back to square one: data extraction.

From my experience, financial institution decision-makers have historically overlooked the extraction component of the business, often because they assume it’s already taken care of. At other times, they accept the current state of play and expect enrichment to fill in the gaps. In reality though, inaccurate data resulting from manual processes remains a core issue across the sector. Without accurate and reliable extraction methods, it’s nearly impossible to automate both intake and future extraction endeavours.

Two things to consider before you make your next tech investment

You can’t bake a cake without preheating your oven or run without learning to walk. And you definitely can’t optimise your data inputs and decision-making without the ability to accurately extract data from disparate sources.

1. Focus on your end users and objectives

While your technology investment should benefit the entire organisation, the needs of employees who will use the technology on a daily basis should come first. Gain a thorough understanding of users’ specific pain points with current technologies and processes, as well as expectations for new solutions. 

Avoid replicating your current process and only changing the technology that facilitates it. Instead, home in on the objective of the process you’re aiming to improve and explore how technology can deliver the anticipated results, without being shackled to the status quo in code.

2. Inventory your expertise (and, where needed, lean on external guidance)

Assemble an internal team to spearhead your technology investment before you greenlight a purchase. Then reflect on the group’s expertise and identify what resources are available to them or are still missing. Do these team members understand the difference between natural language processing and natural language understanding? Can they effectively translate their needs and functional requirements from a solution? If you answer “no” to questions like these, you may want to seek third-party help before making any investments.

If you spend too much time worrying about a solution before understanding your organisation’s data extraction challenges, you can kiss the anticipated ROI goodbye. But with the right expertise and a focus on your most significant pain points, you can pinpoint a technology solution that facilitates accurate, long-term data extraction, and enable time- and cost-saving automation, data enrichment and decision-making.