Custom-Built ERP Module Cures Cash Forecasting Headaches

Heidrick & Struggles is a premier provider of global leadership advisory services and on-demand talent solutions that serves the senior-level talent and consulting needs of the world’s top organizations. “We help companies recruit board members, CFOs, CIO, CEOs, and other high-level talent,” explains Yogesh Tupe, vice president and treasurer. “We also help clients access professionals on an interim, short-term basis through our on-demand talent group. And our consulting business helps clients to identify and assess potential leaders, as well as to shape their corporate culture.”

The Chicago-based firm has been highly successful across all these facets of executive search. “We have more than 60 offices in over 30 countries, and we serve about 70 percent of the Fortune 1000,” Tupe says. “We have a huge presence in EMEA [Europe, the Middle East, and Africa] and in Asia-Pacific.” As a result, the firm comprises 63 different legal entities.

To manage finances and reporting across these dispersed entities, Heidrick & Struggles uses Oracle Planning & Budgeting Cloud Service (PBCS) applications. Until recently, however, the treasury staff of four handled cash flow forecasting manually. “With each monthly close, the team would go into our ERP [enterprise resource planning] system and run a bunch of queries and reports,” Tupe says. “They would export the data into Excel, where they would slice and dice it, and would use the received quarterly FP&A [financial planning and analysis] forecasts to generate each business unit’s cash flow estimates.”

The treasury team would paste all this data into a cash forecasting template, looking for issues with the data or formulas. Every month, this process involved updating more than 2,000 cells. Putting all this data together for the management and operations teams took treasury employees 9 to 12 days. Worse, manually downloading, analyzing, copying, pasting, and reformatting data in more than 500 files per month exponentially increased the risk of errors in the cash forecasts.

To identify potential problems, the treasury team would download bank reports and compare actual account balances with the forecasts they had generated. “Trying to explain any variances caused a lot of headaches,” Tupe says. “A variance might be our mistake—like if we didn’t pick up a particular transaction. There might be an error in a formula. Or it might be a matter of regional differences in reporting. For example, one region might designate a certain line item as a professional expense while another region considered it an office expense.

“We knew about some geographical mismatches, such as that VAT [value-added tax] in some entities was part of a particular expense line, while in other entities it had its own line item,” he continues. “Our team would manually account for those types of issues and would try to accommodate differences in teams’ approaches and perspectives. It was a huge amount of work, and they were miserable for a couple of weeks every month sorting through all the cash flow details.”

Tupe explains that when the treasury group stepped back and reflected on the forecasting process, “we wanted to know whether there was a system that could do all these things for us. Treasury is a lot of fun because you get to tackle problems that are unique and challenging. But when my team was spending all their time on manually manipulating data in spreadsheets, that was pulling them away from the fun part. We wanted to have a computer do the repetitive tasks.”

His team collaborated with Heidrick & Struggles’ IT and financial systems departments to evaluate the organization’s cash flow forecasting needs. This core project team also collaborated with other areas of finance, payroll, FP&A, controllership, and receivables. “For each item in the cash flow forecast, we wanted to understand why it was needed in the forecast and where it came from,” Tupe says. “What were the processes behind that number? What group had ownership of it?” Ultimately, they developed a list of actions that they required from a new cash flow forecasting system.

“One of the big things was that the system would need to convert revenue to collections, which can get complicated,” he says. “For line items like salaries or rent, FP&A data works for treasury because the FP&A and cash flows are both happening on a monthly basis. But for certain items—annual bonuses, insurance payments, or quarterly tax payments, for example—the FP&A is on an accrual basis, so we need to translate it to a cash basis. Previously, these conversions were just based on our judgment, but we wanted to automate all that.”

The project team evaluated a variety of options and decided that, to get all the functionality their cash flow forecasts demanded, they should build a custom module for their ERP system. As ideas for the module crystallized, a member of Heidrick & Struggles’ IT team offered to take a stab at coding it. “The initial results were really impressive,” Tupe reports. “We looked at some prepackaged solutions, but they all had issues and none would fully meet our needs. So we decided to move forward with the idea of the custom solution.”

Designing the ERP module in-house not only ensured it would be tailored to the treasury group’s very specific needs, but also provided a new perspective on cash forecasting for everybody involved. “We gained new insights into our flows—how the flows between all the groups happen, what information they use, and what information they provide,” Tupe says. “In hindsight, I’m really glad we didn’t jump to a third-party solution, because we probably would not have asked the same questions and developed the same level of understanding of our cash flows.”

Senior leadership was very supportive of the software development initiative. The project team focused on quick wins up front, to build trust with the rest of the company that the project was worthwhile. As they demonstrated efficiencies and began to show the lifting of the spreadsheet burden to the different teams, the project’s value became apparent to everyone.

See also:

• Silver Award—Casey’s General Stores: “Unlocking Efficiency in Cash Reconciliations”
• Bronze Award—Hilton: “How Hilton Mitigated the Pain of Bank Fee Analysis”
• The on-demand webcast celebrating all this year’s Alexander Hamilton Award winners in the Technology Excellence category.

The IT team used SQL Server to build out a new module that draws information from the ERP system and interfaces with other internal solutions and with most of the firm’s banks. It pulls cash flow–related information from all these different sources into a forecasting template, mapping categories of flows to specific descriptions, accounts, etc. The module includes complex business logic to account for issues such as seasonality, foreign exchange (FX) fluctuations, and bad debt write-offs.

“We put in a methodology that looks at the last five years of revenue and where we ended up on accounts receivable,” Tupe says. “It also looks at collections each month for the past five years. Seasonality is a major factor in our cash flows; our collections are typically higher in November and December, compared with the early part of the year. So we made sure the system could see what that seasonality looks like, month over month, for a particular year.”

Custom rules also determine how infrequent transactions should be incorporated into cash flow forecasts. And the system takes FX into consideration when generating forecasts in an entity’s functional currency, then calculates the appropriate FX re-valuation when it generates the same forecasts in U.S. dollars, whether for the entity or for rollup into a consolidated view.

Another challenge the module has tackled revolves around the fact that some of Heidrick & Struggles’ banks do not have the capability to participate in automated data feeds to the firm’s ERP system. “Government regulations in certain places require us to use local banks,” Tupe explains. As a workaround, these banks’ account activity is entered into the custom ERP module via journal entries. The team posting those journal entries includes a specific identifier in the Journal ID field, then the module’s logic looks for this identifier and automatically retrieves those cash-related transactions.

The module’s automation reduces both maintenance costs and errors, as fewer functional teams and much less manual effort are needed to support the back-end processes. “Our cash forecasting no longer involves linking, formulas, or guessing,” Tupe says. Cash flow forecasting, which previously took up to 12 days per month, now completes in less than 15 minutes. This frees up a great deal of treasury staff time for more valuable work.

Tupe says that communicating freely and clearly was key to the initiative’s success. “You cannot leave people second-guessing you,” he says. “If you respectfully disagree, that’s fine. And sometimes a person from outside the finance and risk groups might be able to show you a forest that you aren’t seeing because you are so focused on the trees. Keeping an open mind in all communications is crucial.”

In fact, he concludes, such cross-functional communication can benefit everyone involved: “We learned a lot about one another’s areas. Whether treasury was learning about IT, or our IT people were learning about different areas of finance, everyone on the project team was there to learn and solve our forecasting challenges. The innovative use of technology enabled us to transform financial operations in a practical and impactful way, and we accomplished this by harnessing internal expertise across our IT, financial systems, treasury, and finance teams.”