Diving Deeper

Make Big Data Smart Data (Part 1)

Part 1: What’s the Forecast?

“Regarding the weather, so much is said about it, but so little is done.”

That familiar quip comes to mind as I think about data management in today’s financial services environment. An industry statistic that has received about as much play as the comment regarding weather is startling: while 90 percent of global data was created in the past five years, only 1 percent of it has been analyzed. The rest? It’s siloed into databases and file systems, hopefully secure, holding the secrets to critical insights that could translate into more profitable products, better customer service, and more efficient operations.

There are several reasons why this big data hasn’t been managed and structured into models that can create business value. The process can be slow. After all, the amount of data is massive and varied and includes everything from traditional numeric data to unstructured text documents and financial transactions.

In financial services, this process is also complex. Credit card debt is well over $1T; consumer debt, according to the Federal Reserve, exceeds $4T. The velocity at which these transactions create data is astounding. That transactional data doesn’t include, for instance, economic data, customer application data, payment information, and operational metrics.

And then there’s cost, both in terms of people and infrastructure. Leaders with vision know, however, the challenge of cost is offset by the outcomes organizations might enjoy when smart, intentional data management becomes integral to their long-term strategy.

So how can you make big data smart data? What can you do to make sure your organization implements an effective data management strategy to establish a competitive edge?

One thing you can do is to assess your foundational readiness. Just because an organization has decided it’s ready to invest in data management and begin to capture the advantages it offers doesn’t mean it’s equipped to do so. You should also demand expert pragmatism from your consultants. Many of them are classical data engineers – they’re smart – but they lack the real-world experience that comes from having worked in your space.

While data management is certainly an issue that, on the surface, affects the data analytics side of your business operation, you are focused on outcomes: profitability and growth, securing new customers, extending credit to drive value. Your data management team should understand the broader parameters of your organization. They should encourage hands-on input from you to create a solution that is more innovative and effective than what in-house, on-the-ground analysts can likely provide.

I’ll discuss these and other “how tos” in part two of this blog post. But first, let me set the table for that discussion by answering a few critical questions about data management. What does the current landscape look like for data management and how will it evolve over the next few years? Given this landscape, why should financial services companies make data management a top priority?

The Data Management Landscape for Financial Services

As I mentioned at the start, everyone is collecting immense amounts of data. They are essentially working with similar data elements, and they’re applying the same sort of variables from places such as Equifax. The difference between the “have nots” and the “haves,” is that the “haves” are engaged in serious data transformation. The winners in this space have basically centralized data management and the storage ecosystem function. So instead of having one data warehouse for auto loans, another for credit cards, a third for yet another product – each with its own team of analysts – the “haves” created a standardized process across all products and domains. Standardized data is clean and easy to automate, giving you synergy in your processes because there’s no arbitrary uniqueness. Standardization also provides analysts with greater access to data because there’s no learning curve or tribal knowledge. The result? You can turn data into insights and insights into action much faster.

Another prevalent feature on the data management landscape relates partly to security. If you don’t know what data you have and where it is, you’re going to have trouble managing risk. For instance, do you have nonpublic personal information (NPI) and payment card industry (PCI) data living in places that are not carefully locked down? Is that data tokenized? Who’s using it? The California Consumer Privacy Act and legislation like it indicate why it is so important to manage customer data effectively if you intend to maintain trust. You’ll need to be able to tell your consumers where their personal information is on your servers and how it is used. If you don’t know where that data is, you have a problem. So do your customers.

Lineage to source is another aspect on data management that ties into risk. Suppose you’re giving a presentation to the C-suite on profitability of a particular program. If you don’t know the source of your data – after all, it may have come to you through ten different analyst teams – can you be sure that profits were down 5 percent last month or up 3 percent? How many people touched those Excel formulas? There is no lineage to that visibility. As such, you don’t know how your metrics were calculated or where they came from.

Quality, governance, metadata, and other facets also play a critical role when it comes to managing your data effectively.

With robust data management and good lineage practices, you know the data sources and how results were calculated. These results are transparent and defendable – to your C-suite and Board, auditors and regulators. So at the end of the day, it’s not simply a matter of knowing where to find the data. It’s a matter of whether you and the groups that depend on it to make decisions can trust the accuracy.

Making Data Management a Top Priority

Given the current landscape, data management should be a top priority for financial services organizations that want to maintain a competitive edge in a rapidly disrupted market.

One urgent reason for doing so is data risk. A data breach at one credit bureau in 2017 exposed the personal information of 147 million people, resulting in a settlement between $575 million and $700 million with the Federal Trade Commission. In 2019, when a bank failed to effectively assess risk prior to its migration of information technology information to the cloud, a hacker gained access to more than 100 million accounts. That resulted in an $80 million settlement.

The future is private. Most organizations know they’re moving toward a hybrid cloud model or to a new infrastructure. If the powers in charge of that transformation don’t know the terrain, it’s very easy and expensive – both financially and in terms of corporate reputation – to fail in the directive of managing data risk. Imagine the challenge of moving from having a data center on site under lock and key to establishing a virtual presence on a private cloud. If you migrate all your current data to the cloud before you have full control over it, you’ll continue to have legacy data issue. The key difference is that they’ll be located on new and unfamiliar platforms.

There’s another reason why it’s urgent that financial services organizations commit themselves to data management: enhancing profitability. Tools such as machine learning have become democratized, lowering the barrier for organizations both large and small to implement this strategy. Going forward, the players in this space who do exceptionally well and build a competitive edge are the ones who not only manage and understand their data, but also derive actionable, game-changing insights from it. Yes, organizations generate data at a rapid pace. The volume is immense. The content is varied. But it’s imperative that you keep up. When you’re up to speed with your data, your organization is quick and agile. You can see change points on the horizon much earlier because you can run data in real-time, daily, or monthly as opposed to taking four months to run a quarterly process. There’s no reason or excuse for being three months late to knowing a recession started.

The requirement that your organization manage its data with a greater sense of urgency is only going to intensify in the years ahead. Your competitors are working vigorously to execute an effective data management strategy – one that will help them enhance profitability, build new products, minimize exposure to risk, and deepen customer relationships. And today’s small banks and startup FinTechs are on the move, too, not hamstrung by twenty years of infrastructure accumulation. They will grow quickly because they are nimble. Your competition for market share depends on how well you take your bank from being a legacy organization to an information-based, data-driven enterprise. How do you do that?

Stay tuned for Part 2.