Learn how you can Increase cash flow, improve worklist prioritization and reduce DSO through AI Powered Collections Management Solution
Collections management has evolved with the introduction of new technologies. In the past, collections teams manually handled many aspects of account receivable management. The tasks involved sending invoices and statements, prioritizing accounts and tracking payment behavior. They also tracked collections using multiple spreadsheets. Today the collections process is largely automated, with teams increasingly relying on A/R tools to manage their accounts receivable. A collections team’s primary objective is to collect revenue from customers promptly. To achieve this, the team should have these 3 constituents:
Advanced technologies like artificial intelligence and machine learning can help achieve these constituents. They are revolutionizing the industry by allowing many manual tasks to be automated. The future of collections is to automate critical aspects of the process, such as sending correspondence, tracking targets and customers’ payment behavior. This will fundamentally change the way collections is managed and give collectors more time to work on building relationships with their customers and stakeholders to improve business outcomes.
In the course of our collaboration with 500+ A/R professionals, we identified a general lack of awareness and skepticism about the potential that technology presents for resolving the core collections problems. This eBook aims to provide insights into the latest technological advances in collections and how A/R leaders can solve current problems encountered by most collections teams.
We’ll explore the below collections capabilities and their benefits which can be powered by latest technology:
Read this chapter if:
Read on to find out how segmentation based on dynamic factors can help you cluster customer accounts for effective collections.
Traditionally, the customers are segregated into logical buckets based on static factors such as preferred mode of payment, size of the company, preferred mode of the invoice. This enables the collectors to handle each customer segment differently. With business expansion, the collections team has to deal with an exponential increase in the number of customers. There is a need to become more agile in terms of customer segmentation, collaboration and collection strategies.
Customer segmentation powered by AI and ML algorithms such as K-Means cluster will help collections teams take dynamic factors into account while clustering customers. The dynamic factors could be aging, past-due amount, invoice and payment patterns and other relevant dynamic factors for collections.
Let’s now look at a few examples of dynamic segmentation dimensions:
Every customer stands differently in terms of financial stature, market trends and business relation with the company and the collection efforts. For instance, a customer with a double-digit growth rate has evolved from a medium risk to a low risk. The collections strategy for this customer should reflect the leniency in prompt follow-ups and less frequent reminders.
Read this chapter if:
Read on to find out how artificial intelligence and cloud-based machine learning can help you have an effective and efficient collections process and mitigate risks.
What does a day in the life of a collector look like?
Collections agents must do many operational tasks in order to serve customers. These tasks include responding to inbound email, making calls, setting up review meetings and sending out summary notes following a call. They spend a considerable amount of time on preparation for a call. Collectors usually spend around 20% of their time on follow-up activities such as taking notes on disputes, escalations made in calls or emails and tracking down P2P transactions. This results in collectors spending less time in customer engagement.
How is the experience of a collector with AI-powered collections?
AI-enabled collections will streamline the process of collecting payments. Collectors will be presented with a prioritized list of tasks, highlighting the actions that need to be taken. Critical tasks and automatic follow-up reminders will be provided.
An efficient collections software should use business rules and prioritize work lists based on factors such as AI-enabled predicted payment date, amount past due, credit limit, days past due, overall balance, status codes, reason codes.
Using this information, collectors will see a dynamically prioritized worklist and recommend strategies. The following are the recommended strategies formulated based on the factors:
Key benefits of prioritized worklist:
Read this chapter if:
Read on to find out how artificial intelligence and cloud-based machine learning algorithms can help collections predict customer payment dates.
Every day, your collectors reach out to 100+ customers to recover past dues. They typically look at Average Delinquent Days (ADD) metric to prioritize collection calls. This is a reactive approach as they follow up on outstanding payments. Collectors treat every customer the same way as they lack real-time visibility into at-risk customer portfolios. It takes months or even years to understand customer payment patterns and behaviors, which are usually not tracked anywhere. Over a period of time, a few collectors might predict customer payment patterns as they gain experience. If the collector leaves or retires, the organization may lose the information because it is not documented.
A proactive approach towards collections is necessary. It is essential to understand customers’ payment behavior and patterns in order to prioritize those likely to default on their payments. AI and ML algorithms such as Random Forest Regression and Gradient Tree Ensembles will enable collectors to proactively analyze whether customers will honor their upcoming payment commitment or not based on their historical promise to pay trends. Additionally, with AI-recommended next steps, your collectors can customize their dunning strategy for at-risk customers.
AI-powered payment behavior prediction takes inputs such as customer attributes, A/R history and payment history. The data is run through the latest AI model to categorize customers in various predictability buckets.
It will require a minimum of 6 months of closed A/R data on which the AI model can run and provide valuable insights.
Here is the snapshot of the inputs required and the benefits of the AI-based payment model. Payment predictions can have an accuracy of more than 95%. The predicted payment date can be leveraged by changing the collections strategy.
Read this chapter if:
Read on to find out how artificial intelligence and cloud-based machine learning can help you identify intent from customer’s emails and scale up dunning outreach with automated emails.
Automated intent identification for incoming emails
Receivables managers and analysts process hundreds of thousands of transactions every day, looking for data, information, and reports across systems and tools. Let’s take the example of Lara, a collections analyst. When she logs in for the day, she finds a lot of incoming customer emails. The request could be to identify and match an invoice or account statement. Customers routinely request additional invoice copies, bills of lading, and proof of delivery, among other invoice and order-related documentation. This requires pulling data from multiple resources and performing time-consuming analysis.
It is essential to understand the customer’s intent and revert to their requests quickly so they can be serviced effectively. A solution has to be sought that brings all the necessary information to a single place for collections analysts to access and serve.
AI and Natural Language Processing(NLP) can parse incoming emails from customers and bring them to the worklist. With AI, you will see action items such as creating P2Ps and disputes. It will identify what customers are trying to say and recommend actions for Lara to work upon. The system will process all these actions, and Lara will see the emails in her inbox and the recommended steps. Today she has to create 5 P2Ps, 10 disputes which she has to record, etc. All Lara has to do is click on the action and send the invoice copy, with a pre-drafted email for your response.
The figure above shows the current and future state of collections. The left side indicates the current state, where collectors have to perform manual tasks. The right side indicates how automation will improve each task and how collectors’ jobs will become more efficient as a result.
Automated dunning and interactive outgoing emails
Collections as a process have never been an easy one. However, the current
economic situation has made it even more difficult. Most customers ignore repeated
reminder emails and delay payments over the due date because of cash flow issues.
As a result, most mid-sized businesses see a rise in DSO and a downfall in overall cash
To streamline the collection process, a correspondence automation solution is essential. It will handle and manage multiple templates for correspondence. It should also support mediums such as print, email, fax and include data about phone-based touchpoints (e.g., dunnings). Furthermore, it will help automate manual tasks involved in creating correspondence, such as selecting accounts and contacts and sending out emails or dunning notices.
Manual drafting of dunning emails for customers is a time-consuming task. Collectors cannot prioritize at-risk customers, resulting in a ‘same size fits all’ strategy for all the customers. Using AI, collectors could enable touchless dunning for low-risk customers. They can send out bulk correspondences with the help of a single click. This will enable collectors to focus more on critical customers.
Additionally, by leveraging AI, collectors can send emails at the right time. When an email is sent during busy hours, there is a high likelihood that customers might miss it. To improve readability and effectiveness, collectors can use AI to:
The system will consider all these factors before sending the email at the right time. When you send the email, the system will not send it in real-time but will queue it up and deliver it at the most appropriate time, improving the chances of your email being read.
Read this chapter if:
Read on to find out how artificial intelligence and natural language processing will enable efficient inbound and outbound customer call management
Let’s say Lara has a discussion with a customer on credit limit extension. To cater to this request she has to look real-time at customer payment history and total credit exposure. This requires pulling data from multiple resources and performing time-consuming analysis. Similarly, a typical day for any order-to-cash team involves countless hours lost in looking for information.
With the help of artificial intelligence, Lara can see how close she is to reaching her collection goals. The system will prioritize her worklist, keeping in mind what she has already accomplished and what more she has to do. How many customers does she need to contact today? Who does she need to contact? Leveraging artificial intelligence, we can do a lot.
Now that Lara has to reach out to people via call, let us see how we can leverage AI to assist her with her task. An AI-powered virtual assistant similar to Siri or Alexa would make the lives of analysts like Lara easier. Users can perform all collections tasks with an AI assistant by giving a voice command. It would automate tasks such as calling the customer, taking live notes or creating action items with just a voice command. The AI assistant will organize your day by prioritizing customers to call, emails to send, and recommended strategies based on previous interaction history, notes, and P2Ps.
The assistant will make the analyst’s work more efficient by automating tasks such as collecting call logs, creating summary emails and recommending actions such as P2P creation. Additionally, it will summarize why a customer is on the call list today and eliminates time spent during preparation for calls. The AI assistant will bring up a page where you can see why the customer is on the call list today, total due, current due, past due, insights from the last few communications made with the customer, credit utilization, portfolio summary etc. It also captures call notes and summarizes next action items from calls in an email format.
Read this chapter if:
Read on to find out how Webcrawler API and RPA Integration can help to provide the latest payment status information and promise to pay extracted from respective customer AP portals against your open invoices.
Customers often post updates on open invoices to their portals. Collectors must monitor the promise to pay commitments by customers. It is a time-consuming process as collectors will have to log in to each portal and search for specific invoices’ status. The team may have to spend hours pulling in information from portals and duplicating the data by copying the same information in notes.
With RPA-enabled invoice status tracking functionality, 50-70% of these activities will be automated. Collectors could automatically get any payment status information posted on customer payment portals against respective invoices.
This would result in reducing the need for follow-ups with customers and improves the effectiveness of collections processes. The system can also automate the creation of notes, promises to pay, and disputes based on the information from the portal.
With payment tracking automation, collectors could monitor and record a promise to pay. The system can alert the collector if a P2P is broken. Further AI can also be leveraged to predict whether a customer-created promise to pay will be honored or broken based on a historical commitment of keeping trends. The payment commitments are also evaluated to be kept, partially kept, or broken based on the payments received and marked appropriately.
Read this chapter if:
Read on to find out how seamless integration can help to resolve disputes to fast-track resolution and reduce workload.
Large enterprises often have separate collection and deduction teams. Collectors usually aren’t equipped with the right tools or transparency to look into the disputes raised by their customers. The lack of real-time visibility or proper access to disputes hinders the process of dispute resolution, but integration can improve this process.
With AI, an automated solution will facilitate streamlining collections and disputes. An integrated feature, such as a disputes tab, can let collectors access all the customer disputes from one place. Collections specialists can expedite various processes with
this feature. This can expedite various processes for tasks such as working on a range of disputes and corresponding with relevant members to routing them to different teams for quick dispute resolutions.
Here is are a few tips on how dispute management can be simplified or automated
Collaborative electronic workflows will streamline the resolution approval across various stakeholders based on the internal hierarchy of organizations. Deduction analysts can approve, reject, or reassign deductions to another stakeholder with a single click. This way, deduction analysts and stakeholders gain end-to-end visibility on every deduction through a centralized repository of notes.
Collections is critical. Hence you need to enhance their capabilities to have A/R operations produce the best outcomes. AI, ML, RPA and other latest technologies are game-changing enhancements. They take A/R processes and departments from functional to world-class.
Financially, it’s incredibly costly for companies to build such solutions in-house. Startup and accrued cost from continual maintenance dissuade many businesses from exploring the power of software and automation.
Another option is to partner with a vendor who can seamlessly add automation and intelligence to your A/R operations without the burdensome overhead. Schedule a demo with one of our A/R specialists to learn how Autonomous Receivables can take your enterprise to new heights.
Automate invoicing, collections, deduction, and credit risk management with our AI-powered AR suite and experience enhanced cash flow and lower DSO & bad debtTalk to our experts
HighRadius Integrated Receivables Software Platform is the world’s only end-to-end accounts receivable software platform to lower DSO and bad-debt, automate cash posting, speed-up collections, and dispute resolution, and improve team productivity. It leverages RivanaTM Artificial Intelligence for Accounts Receivable to convert receivables faster and more effectively by using machine learning for accurate decision making across both credit and receivable processes and also enables suppliers to digitally connect with buyers via the radiusOneTM network, closing the loop from the supplier accounts receivable process to the buyer accounts payable process. Integrated Receivables have been divided into 6 distinct applications: Credit Software, EIPP Software, Cash Application Software, Deductions Software, Collections Software, and ERP Payment Gateway – covering the entire gamut of credit-to-cash.