Account receivables management at companies across a broad range of industries suffers from severe limitations that are leading to avoidable trapping of significant value. Disparate and non-intelligent IT systems, duplication, inaccuracies, inconsistencies in data and information across collections, finance, treasury, operations and customer departments are severely affecting the efficiency.

Intuition-based approaches and loosely defined key performance indicators (KPIs), and use of one-size-fits-all strategies are factors leading to entrapment of value that would otherwise be available for business growth. Reactive processes are inadvertently guiding teams to focus on accounts with poor conversion ratios, and escalations that surface almost around the time that risk events occur, preventing effective management by exception. Limited staffing is another constraint that hurts the efficacy of account receivables functions.

Part 1 of this series provides detailed insights into various challenges facing the account receivables function. This installment offers a solution.

Digital to the accounts receivable rescue

All the above challenges represent fertile grounds for infusing new digital technologies to build a proactive and more efficient account receivables management function. We present a four-pronged digital strategy to effectively manage the account receivables by turning information silos into systems that are symbiotically tied to each other, supplanting one-size-fits-all approaches with tailored strategies, automating follow-ups and receipt reconciliation processes that collectively help reinvent the incumbent systems managing account receivables.

• A decision-support dashboard. Accounts receivable management begins even before the invoice is issued — when the decision to sell to a customer on credit is taken. And it ends with the payment receipt and reconciliation. Machine learning (ML) and analytics technologies should be leveraged to bring all this information under a single dashboard to weed out the inefficiencies caused by disparate systems and information gaps between relevant departments in the company.

  The dashboard acts as a control center empowering the user to perform collection activity-related actions such as capture correspondence logs, send reminders to counterparty by clicking a button, or involve finance team in payment receipts. A system built specifically for management of accounts receivable processes will improve the decision-making process and ensure that collection policies are followed.

• A tailored collection strategy. Collection strategies are often generic and based on reactive parameters such as invoice aging and invoice value. Prioritization of tasks for the receivables collection is often ad hoc and driven by company staffing constraints. This results in a standard collection strategy followed for all customers — which is outdated and ineffective.

  An ML algorithm powered by analytics can suggest customer-specific, tailored collection strategies. Once they’re fed into a regression algorithm, parameters such as average time for payment, invoice value, disputed receivables, customer credit score, and payment terms will give insights into time variation in payment of invoices vis-à-vis the invoice value and percent likelihood of invoice payment based on invoice aging and value.

  A good example would be where a prescriptive algorithm, based on historical behavior and the macroeconomic situation of a customer’s industry, could suggest a dynamic receivable discounting strategy that would result in earlier payment of the invoice.

  Eventually, AI and analytics can be used to parse internal sources such as the customer’s history, behavior and external sources such as financial statements and online media to influence payment terms, credit strategies and collection strategies.

Figure 1

• Automating follow-ups and reminders. Analytics can profile a counterparty based on credit exposure and response to past reminders and follow-ups to identify the right time to initiate follow-up communications and the right person to send it to.

  The ML algorithm can parse the past actions of the user on the dashboard and send automatic payment follow-up mail to low risk-higher payment probability customers, thereby freeing up time for the user to attend to more important accounts.

• Automating of receipt reconciliation. Payment receipts are sent by counterparties in physical form as well as electronic form such as email, spreadsheets and online portals. It can be a daunting task for short-staffed collections teams to reconcile the receipts with invoices.

  Optical character recognition (OCR) and ML can be deployed to automate this process. OCR converts physical receipts to digital format. ML programs can process electronic and OCR-converted physical receipts and match them with payments based on historical matching patterns.

This article was written by Vinod Malpani, Pramit Basu and Naveen Krishnan from Cognizant Consulting’s Banking and Financial Services Practice.

To learn more, visit the AI & Analytics section of our website.