You may also be interested in:


Treasurers, It’s Time to Automate Cash Application

  • By Riddhima Batta
  • Published: 7/12/2018
The fundamental goal of cash application in accounts receivable (A/R) is to efficiently process the incoming payments and post cash. Companies struggle with applying payments due to inconsistent channels, formats and data with payments and remittances.

A survey by NACHA and CRF revealed the payments mix for B2B companies has undergone a drastic shift, in which e-payments are on the rise. However, cash application teams have yet to ride this wave of changing trends as most companies are stuck with traditional lockbox solutions.

Pain points of traditional lockbox solutions

Incomplete support for processing checks and high lockbox fees: High volume of checks necessitates costly lockbox services. Banks charge capture, keystroke and transmission costs. Consequently, companies could end up paying $1 to $3 per check. This directly impacts bottom line.

No support for e-payments: Bank lockboxes do not support remittance reconciliation for e-payments. Given the rise of e-payments, manual processing is a costly and inefficient affair, with ripple effects visible across credit-to-cash cycle.

A/R teams reduced into transaction management: Low-value, manual tasks involved in cash application reduce A/R teams to doing back office, clerical work instead of focusing on customer engagement to improve repeat business and unlock latent business opportunities.

Process automation and AI for cash application

Process automation is the use of robots mimicking human actions to perform well-defined functions. It is best-suited for performing repetitive tasks by following a fixed set of rules. However, it is based on strict compliance of orders and highly structured input, eliminating scope of learning.

Artificial intelligence (AI) is the ability of computers to learn, reason, think and perform tasks requiring complex decision-making. AI can perform tasks by analyzing humans, evolve with experience and looking for better ways to execute and handle newer inputs based on experience.

A/R can leverage process automation to improve straight-through processing by:

Invoice matching using non-standardized data: AI could match payment and remittances with open invoices using non-standardized data like:
    •    Invoice number/account invoice references
    •    Truncated invoice numbers
    •    Alternate references like purchase order numbers, BOL numbers
    •    Alternate payers in case of a complex parent-child structure.

AI-enabled OCR capture: AI-enabled OCR captures remittance information from check stubs using template-free technology. AI not only reads data but understands it as an analyst would, distinguishing different fields such as the invoice number, due date and purchase order number. Eventually, the accuracy increases with self-learning.

Eliminating repetitive exceptions in cash posting: Approximately 80 percent of cash posting exceptions are repetitive in nature because customers repeatedly send remittances with the same inconsistencies. AI-based solutions learn the pattern in which analysts handle exceptions, and auto-resolve them without any analyst input for similar exceptions.

Integrated Receivables

Integrated Receivables is a platform that allows different A/R processes like credit, collections, deductions, cash application, billing, and invoicing to collaborate. To leverage the most from a unified A/R process, the following needs to be optimized.

Core transaction management: In the current scheme of things, analysts spend the majority of their time in low-value work. These low-value activities need to be automated, enabling analysts to focus on strategic tasks, such as researching invalid deductions, controlling credit risk and reducing critically delinquent accounts, rather than just chasing payments long after the transaction happened.

Human decision-making: AI processes large data sets and provides analysts with simplified metrics for better decision-making. Examples of AI facilitating human decision-making are:

Blocked order prediction: AI predicts the likelihood of orders getting blocked by monitoring the credit limit and past purchase patterns of customers. Based on this, analysts take proactive action to update credit limits or communicate with customers preemptively.

Payment date prediction:
Traditional collections management relies on skill and speed of individual collectors and kicks into action after a customer has gone past-due. AI helps collections become proactive by predicting payment dates of an invoice.

Dispute validity prediction: A/R teams often struggle with deductions paradox where they are not able to prioritize their efforts without knowing the validity of each deduction—which remains unknown until research iscomplete. By studying past resolution patterns, current deduction characteristics and applying machine-learning algorithms, AI is able to predict which deductions are invalid, saving time and productivity lost in         researching valid deductions, reducing instances of false-positives and saving dollars written off for invalid deductions.

Structured internal collaboration

Centralized repository: A centralized repository for documentations across processes such as claims, BOL, POD, remittances, credit applications and deal sheets gives stakeholders like collectors, brokers and deductions         analysts quick and easy access.

Standardized workflow: Workflows for cross-department collaboration are predefined for use cases such as blocked order resolution, identifying disputes and dunning customers.

Manager approval escalation: For high-priority tasks, such as approving the credit limit for critical accounts, it is escalated to the manager/director from analysts, within the system.

Simpler external collaboration

Automated mass correspondence:
Emails are sent to customers en masse instead of manually being sent.

Predefined templates: Denial packages and dunning templates are built-in with necessary backup documentation, ready to be sent to customers/vendors.

AI provides finance executives with comprehensive process visibility, driven by real-time integrated data. Advanced reporting enables a consolidated overview of statistics and simplifies decision-making. AI positions management officials to drive next-level process transformation versus policing metrics and process owners.

Riddhima Batta is product marketing manager for HighRadius Corporation.

Copyright © 2021 Association for Financial Professionals, Inc.
All rights reserved.