Print Page | Contact Us | Sign In | Join RVCF
Blog Home All Blogs
Search all posts for:   


View all (430) posts »

Disputes Resolved with Artificial Intelligence

Posted By RCVF Admin, Wednesday, August 7, 2019
Updated: Friday, July 26, 2019

RVCF Link Image 

Disputes Resolved with Artificial Intelligence

By Sonali Nanda, Vice President of Product Management at HighRadius


According to Attain Consulting (1), dispute and deduction analysts spend over 1600 hours trying to analyze claims from big box retailers.

AR teams handling retail customers want to resolve open disputes and clear payments as soon as possible. However, ambiguous communications, unstructured document management, siloed operations combined with the high volume of claims from big box retailers leave analysts wanting for some sort of assistance.

Dispute and Deduction analysts still follow and use traditional processes and strategies to handle disputes and deductions, while other AR functions such as credit management, collections management, cash application, etc. have significantly improved due to automation. In addition, analysts require logging onto retailer/vendor portals to access the claims.

Some of the challenges faced by dispute and deduction analysts especially regarding claims from big-box retailers are:

  • Manage large volumes of payment deductions
  • Coordinate with various teams regarding each dispute/claim
  • Gather all supporting documents for every dispute/claim
  • Spend extended amount of time on analysis

An Overview of How Analysts Handle Big Box Retailer Claims:

  • Login to vendor portal: The lack of push notification forces analysts to login to the portal multiple times to check for claims.
  • Download claims: Vendor portals usually do not support bulk downloading; each claim should be downloaded manually.
  • Upload document: Documents including remittance, POD, BOL, etc. need to be uploaded manually and individually for each claim.
  • Information key-in: Information regarding each dispute should be keyed in manually.
  • Manual correspondence: For each invalid dispute, analysts manually key-in and send denial correspondence along with the supporting documents, which include BOL, POD, claims, etc. to big-box retailers.
  • Cross function collaboration: Analysts often need to interact with other functions within AR for dispute resolution. This only elongates the entire process.

As a consequence, the entire dispute and deduction process is

·       Highly manual

·       Time consuming

So, the question here is, is there a solution that:

·       Automates the process, and requires minimal human intervention

·       Is time efficient

·       Adds significant value to the entire AR functions

The simple answer is Yes! Artificial intelligence can automate the entire end-to-end process and requires minimal human intervention. It is effective, efficient, and adds tremendous value to not just the dispute and deductions process but the entire AR function.

What is Artificial Intelligence in AR?

The leading or general understanding of artificial intelligence is ‘the ability of a machine or a computer program to think and learn on its own.’

Leading organizations around the world have enabled their account receivable teams with artificial intelligence allowing them to effectively perform their day-to-day activities. AR functions such as cash application, credit management, collections management, etc. are among those that use artificial intelligence the most.

However, AI is not limited to just those functions. AI helps dispute and deduction teams as well.

How Does AI Help Analysts Handle Big Box Retailers?

Auto-coding of Trade and Non-Trade Deductions

Solution: Claims and disputes are downloaded from big box vendor portals. They are identified as trade or non-trade deductions using AI. Trade deductions are usually processed in a few seconds; however, non-trade deductions require capture of respective BOL, POD, email and remittance. Reason codes are identified and auto-mapped to internal codes using AI and machine learning.

Benefit: Decrease in time spent on identifying and coding disputes. Direct assignment to analyst for further action.

Dispute Validity Prediction and Higher Recovery Rate

Solution: Taking into consideration, historical deduction data, all dispute related parameters, and factors related to cleared invoices and applying prediction models, the machine learning algorithm can predict the invalidity of a deduction with a high confidence level. This provides a prioritized list (with Deduction Recovery Estimate) of high dollar value invalid deductions for analysts to focus on. In effect, Artificial Intelligence helps analysts achieve quick deduction resolution and a higher recovery rate.

Benefit: Analysts can focus their energy solely on trying to resolve high research claims.

Simply put, Artificial Intelligence streamlines dispute and deduction activities not just for big box retailers, but for deductions, claims and disputes in general.



Sonali Nanda is the Vice President of Product Management at HighRadius, leading the development of the Integrated Receivables Cloud Platform and other product management initiatives, including the home-grown Rivana Machine Learning platform. She is responsible for building an effective product management team and defining product strategies and roadmaps, which accelerate sales and create happy customers.


This post has not been tagged.

Share |
Permalink | Comments (0)
          Innovative Retail Technologies EDI Academy