Engineering a new approach to cash management

What the company needed To regain control of and reduce its past-due receivables and enhance cash flow To improve the cost to serve while bolstering customer satisfaction

The knock-on effects of poor customer data Our client has a century-old tradition of producing industrial products that improve the lives of people around the world. But there was a disconnect at play. Both its equipment and service businesses in North America were experiencing payment delays. The reason: it had a traditional approach to collections and cash management strategies that treated every customer identically. It used dollar thresholds to set collections in motion without a clear understanding of the specific factors influencing the high level of aging invoices. It based collections and credit policies on limited data and historical information that didn't account for evolving payment patterns or customer behavior, making it hard to build a strategy that would reduce past-dues. As a result, days sales outstanding among its North American customers averaged nearly two months. That meant the cost to serve each customer was high, cash flow suffered, and the company risked losing revenue to bad debt.

Revamped collections processes. A predictive model for better customer care Having managed the company's finance services, including order to cash, and provided analytics, procurement, and IT services, we had a good overview of its operations, but then it was time to dig down. Taking a hard look at receivables management We scoured its accounts receivable and enterprise resource planning platforms for data and intelligence on sales, contracts terms, and payments. We performed in-depth analytics and risk assessments on customer behaviors and characteristics We took a detailed look at the company's payment and credit terms, reviewed the products customers bought, examined collection processes, and more We assessed a list of more than 10,000 active customer accounts for delinquency figures and invoice volumes to distinguish the dependable payers from the rest Gathering this data helped us pinpoint factors leading to payment delays. This intelligence allowed us to build a predictive model, powered by advanced analytics, that forecasts which customers are likely to pay late, early, and on time. This model also uses machine learning to help maintain quality collection practices. Having segmented the company's customers, we could identify reliable payers who need little or no follow-up on their past-due or low-balance accounts. We then automated activities that made collecting from them touchless.