Six Big Data Hurdles Payments and Financial Professionals Need to Overcome

Analytics. Big data. These are terms that organizational leadership has been hearing for years. From finding ways to capture more discounts from suppliers to reducing wasteful spending to overcoming fraud, everyone in the payments network is finding use for big data analytics.

With more information that organizations need to process, the phrase big data often feels like an overused buzzword in the technology community—with ambiguous definitions and platitudes to match. But it’s much bigger than that.

What is Big Data?

In its most simple terms, big data is essentially the deluge of information that can be analyzed computationally to reveal patterns, trends, and associations, especially relating to human behavior and interactions.

Such data, available in structured (e.g., columns of numbers) and unstructured (e.g., free-form text) formats, can be harnessed to derive insights that drive understanding of customers, differentiation in value proposition, need-based innovation and top and bottom line growth. Experts attribute the following characteristic features to Big Data:

• Volume: The amount of data to be handled is huge, and is derived from multiple sources – be it structured transactional data or unstructured data – including sources such as social media. Earlier, data storage was the main cause for concern; but with decreasing storage costs, concern has now shifted to integrating with the many sources of data and then deriving value from the data.
• Velocity: Refers to the speed of data generation, availability and processing ability.
• Variety: Data sources can be as diverse as social media and networking websites, CRM software, transactional processing systems and other applications that form a part of the technology infrastructure of an organization.
• Integrity: Data coming from multiple sources becomes usable only when it is cleaned, validated for integrity and correlated to identify relationships, hierarchies and linkages. The complexity of this exercise stems from the need to determine the required mapping to common definitions and bring structured and unstructured data under a common framework. Data inconsistencies due to source environmental variability add to the complexity.

How Big Data Can Fit in Finance and Payments

When translated to payments and finance, big data could mean many things. To answer this, the Association for Financial Professionals recently released a guide for finance professionals looking to make big data analytics work for them, the AFP Guide to Big Data.

From the guide:

“FP&A executives are beginning to realize that by utilizing external, internal, structured and unstructured data, they can greatly improve the quality of forecasting and planning. Better, faster information suits FP&A’s growing role as a true strategic partner. The objective is to give management the information it needs when it needs it, so it can make smarter decisions to boost enterprise performance.”

Challenges to Overcome in Implementing Big Data

With so many opportunities to use big data, there exist challenges including a talent gap, a need for a shifting mindset, new ways of measurement, data ownership, visualization, and finding a starting point.

Missing Skills

When it comes to big data analytics, finance faces an uphill battle to achieve the training and expertise needed to make all of this data work.

According to FP&A consultant Jenny Okonkwo,

“Big data calls for different competencies; candidates need to be very comfortable around manipulating, extracting and analyzing data to draw insights. This is essential to get the right story, leading to the right decisions being made at the right time.”

To overcome this, finance training programs will need to incorporate data analytics and insight management, preparing professionals for dealing with data discovery and high-level analytics.

Dealing with the Unknown

Finance is a straightforward field. Numbers historically needed to be able to stand up to intense scrutiny and generally accepted principles—but big data deals with many unknowns.

Roger Fried, senior data scientist at Teradata Aster, and a former finance director contends that finance has grown up with an arcane way of “bucketizing” everything, the result of a lot of old laws and regulations.

“This segmentation frustrates the use of statistics,” he said. “We aggregate for simplicity.” By doing so, “we’ve already ‘garbaged’ up the data with wrong assumptions, which hurts the process of squeezing information out of data.” That leaves finance with a set of small vs. big data — the data that’s been traditionally built into the language of business and has allowed for common conversations. “That attitude really limits how finance can apply big data,” he said.

To overcome this, finance will need to think creatively and fit this broad thinking into the current governance structure.

A Need for a ‘Cultural Evolution’

Dealing with the unknowns, as talked about in the prior point, is part of what makes the culture of finance risk averse:

“There’s a degree of concern that leads to obfuscation,” said Vic Datta, CEO of the business management consultant Resilicore. “Managers don’t want to put their numbers against external performance measures and competitive information, especially when they are not certain about how to consistently get these new measures or how to address them.”

To successfully adopt big-data strategies, there must be a cultural shift in organizations, led by finance to allow performance dips to be tolerated for a period of time in areas where new measures are being developed so that performance is always being measured by the right indicator and benchmark.

Whose Data is It?

An issue, especially for decentralized companies, is the issue of data ownership. Without ownership and accountability, there won’t be a single area that focuses on adding value to the enterprise and optimizing the results for the company as a whole over its own individual business unit.

Visualization Tools

In terms of big data, finance lacks other departments in terms of visualization. For instance, marketing has made amazing use of trends and information, creating visualizations for the purpose of generalizing sales. Finance, on the other hand, is tied by spreadsheets and remains behind the curve in visualization adoption. Our first data analytics product offering at WEX, ClearView, has focused on data visualization, providing users with an easy to understand view of the data as well as the ability to interact with data. We have found that visualizing and understanding data is a critical first step before users are willing to take action based on data-driven insights.

Finding a Starting Point

A part of big data is the sheer size of the information being processed—leading to the question, “where do we start?”

Using tools that provide a GUI into large data sets is not enough if the user doesn’t have business outcomes in mind and a well-developed project management plan. The results need to be practical and actionable to have any value. Roger Fried, senior data scientist at Teradata Aster, believes that Excel for finance is an individual effort, and “you need to make advanced analytics a team sport.”

Making Big Data Work for Finance and Payments

Whether in FP&A, payments, accounts payable, or any other department, being able to take in all of the information available to you in order to analyze trends is big business. Gain end to end visibility, streamline payments, save money and more.

Learn more about putting big data to work in “Unlock a Treasure Trove of Insights into AP Data,” and read through TATA Consulting Services’ whitepaper, “Big Data in Payments – Unparalleled Opportunity for Strategic Excellence” to learn about bringing effective analytics into your payments strategy.