Sink or Swim: How Data Lakes Remove Silos and Improve Cash Visibility

FinLync | June 6, 2022

In a recent webinar, experts from FinLync and Actualize Consulting discussed how treasury teams can use data repositories like data lakes or warehouses to increase cash visibility and remove siloes across teams. Though treasury teams could certainly spearhead the creation of a data repository at their company, leveraging an existing data repository that has already been established is a far easier approach. 

Before we begin, an important note: there are several different types of data repositories, the most common being data lakes and data warehouses. We have used these terms interchangeably within this article, however there are plenty of other options available today. To better understand how these differ, we recommend referencing The Treasurer’s Guide to Data Lakes. 

What problems do data lakes solve? 

Decisions made by treasury can only be as sound as the data they depend on. Yet, treasury’s data sources are often stale and disconnected. Treasury teams and their partners in IT have tried to solve this common headache in a variety of ways: work-around bridges and connectors between systems are vulnerable to hacking, are prone to breaking, and inconsistencies in format between systems demand laborious cut-and-paste work for treasury, inhibiting daily progress. 

Chad Wekelo, Principal at Actualize Consulting, “If I’m a typical organization, I have data all over the place. I have disparate data in multiple places. No one source is going to have everything I need.” 

Rather than trying to connect disparate systems to each other, data repositories offer a way to master data from multiple systems and sources by connecting each system to one central “sink” of data. Each different system is a faucet and all the data flows into a shared sink – your data lake or data warehouse. 

Get the free ebook: The Treasurers Guide to Data Lakes

What broad problems do data repositories solve for treasury? 

First, data repositories tear down the walls between internal systems, and the modules within those systems. Now you do have one single source for all the data you need. For treasury, this means a dramatic decrease or potentially eliminating all the manual copy/paste work that is done on a daily, weekly or monthly basis and being able to discover data that treasury has never had access to before. 

Second, according to Robert Granger, Senior Manager at Actualize Consulting, “It really lets you ask much more complex questions because that data repository has a much more diverse array of data within it.” Granger continues “You might need to start applying statistical techniques, extrapolating trends on forecasts, or spotting correlations with payment fraud. But to boil it all down, that data lake’s going to provide that holistic view that you might not be able to get from a single system to help you make better decisions in a timely way.”  

The Anatomy of a Data Lake 

 

Specific Treasury Use Cases for Data Lakes 

Cash Visibility: Wekelo shares that with a data lake, more data is available for your longer term forecasting, which means the opportunity to analyze trends, behaviors and even introducing predictive analytics and AI.  For example, if treasury can access all payment history for all vendors, they can combine some of the financial transactions, do covenant reporting and disclosure reporting, enabling more working capital analysis to be done. This capability is above and beyond what can be done in a treasury workstation, because the TMS doesn’t have that level of granularity.  

Cash Forecasting: Mitch Thomas, Head of Solutions Engineering at FinLync comments, “Because data lakes are using the most up to date source for financial contracts that are outstanding between an organization and its customers and its vendors means treasury can how those payments were made. And more importantly, how those receipts were received.” He continues, “So you take not just the information in that ERP, where you’ve got 30, 60, 90 days worth of actual financial documents posted for payables and receivables, but you’re able to take that history and the activities that have been made to be able to predict out into the future what your payables will be, what your receivables will be, with a high level of accuracy, because you’ve brought in all of that information from the ERPs from a not just an invoice perspective, but from a clearing perspective. And that allows you to do some really good forecasting when it comes to things like best case, worst case, most likely.” A significant advantage for both treasury and FP&A. 

Dynamic Discounting: With access to this breadth of information, such as trends, the payments, the invoices, even some of the receivables they have, it allows treasury to make strategies for utilizing that cash. It opens the door to a dynamic discounting program for certain vendors because treasury can see that they’re always paying on a more extended period, so maybe the vendor needs more cash, says Wekelo. 

Accessing Bank Data in (true) Real-Time: Data lakes are setup to easily connect to APIs. The easiest way to add another “faucet” of data to your repository “sink” is by plugging in the API to the data repository. Now that corporate bank connectivity has advanced beyond host-to-host and file-based connectivity to bank APIs, companies can finally have the up-to-the-second bank data instead of relying on the bank to send them statements.   

Optimizing Disbursement Mix: Wekelo also suggests that the use of a data lake means treasury can analyze the different types of disbursements that are being used. Are there a high quantity of checks or wires? Can treasury be smarter about the mix of those, and even optimize for the fastest payment type? 

“I think the main reason treasury is not thinking this way today is because they just don’t have access to the information,” concludes Wekelo. This is precisely where a data repository can assist. 

Get the free ebook: The Treasurers Guide to Data Lakes

Who else can benefit from a data repository? 

One could make the case that every department can benefit from a company-wise data lake or data warehouse. This is exactly why so many companies have spent the time and money to establish them. 

For teams that belong to the office of the CFO, the beneficiaries include: 

  1. Accounts Payable: increases efficiency of both the outbound payment process and monitoring efficiencies across the organization thanks to having all data in one central place 
  2. Accounts Receivable: easily consuming information across several systems including bank data means being able to clear items more quickly and easily 
  3. FP&A: enhanced analysis is made faster and easier for things like forecasting 

 

The IT team also sees a major benefit with data repositories. Establishing and maintaining countless connections among systems, all of which are prone to breaking, is highly time consuming. Troubleshooting is challenging because there are so many points at which different systems and connections can break.  Companies with a data repository focus their efforts on a simpler hub-and-spoke of the data lake and the “faucets” that feed into it. 

 

What does a day-in-the-life of treasury look like when using a data lake? 

Consider ascenario where the data lake stores all financial information for the organization, including real-time, up-to-the-minute bank data from all your banks, and all finance data from all the subledgers including today and into the future. In this situation, the treasury practitioner’s day looks like this: 

According to Mitch Thomas, Head of Solutions Engineering at FinLync, “We’re starting off with a single data source where all the information has already been compiled. I already know at the start of the day what potential flows are going to impact my treasury operations for the day. “With a data repository connected to your banks via API connectivity, today’s real-time, up-to-the-second balances are delivered at the click of a button. Treasury teams can refresh bank data whenever desired. This contrasts with most cases where the team is working off a previous day balance.  

Thomas continues “Instead of, in most cases, that data getting less accurate throughout the day, it’s going to get more accurate because that data lake is going to continuously be updated with data from all the connected systems, along with the updated bank information. So you’ll see a lot of that information that was in a forecasted bucket at the beginning of the day, where these were projected inflows or outflows. Throughout the day, you’ll see those move into the actualized bucket and get a clearer and clearer picture of your cash throughout the day, as those payments are debited or credited at your bank.” 

The result 

“Treasurers can be tighter with their cash, allowing them to go out and invest more of those funds and borrow less, which, as our interest rates are trending in that direction, becomes more and more important. What we really see is that data lake enabling treasury to really be nimble because they can use that comprehensive, updated data set to make the most-precise decisions,” says Thomas. 

 

Data Lakes