Company: IBM, Armonk, NY
Company Description: IBM is the hybrid cloud and AI technology and services company, focused on providing client value through a combination of technology and business expertise. IBM solutions draw from an industry-leading portfolio of capabilities in software, consulting services and a deep incumbency in mission-critical systems, all bolstered by one of the world’s leading research organizations.
Nomination Category: New Product & Service Awards Categories
Nomination Sub Category: Sales Enablement Solution – New
 Nomination Title: My AR InBox and Virtual Assistant - Driving Speed and Accuracy in Accounts Receivable
- Which will you submit for your nomination in this category, a video of up to five (5) minutes, explaining the features, functions, benefits, and results to date of the nominated product or service, OR written answers to the questions? (Choose one):
Written answers to the questions
- If you are submitting a video of up to five (5) minutes in length, provide the URL of the video here, OR attach it to your entry via the "Add Attachments, Videos, or Links to This Entry" link above, through which you may also upload a copy of your video. If you are submitting written answers to the questions for this category, provide them in the spaces below.
- Reference any attachments of supporting materials throughout this nomination and how they provide evidence of the claims you have made in this nomination (up to 250 words):
- Describe the features, functions, and benefits of the nominated product or service (up to 350 words):
Total 343 words used.
My AR Inbox module is a new AR email service embedded into IBM’s Strategic AR System EngageAR). This solution has transformed how our AR professionals communicate with our clients and goes above and beyond what traditional email platforms provide. My AR Inbox allows for client responses to be automatically embedded into the relevant invoice(s) allowing an end-to-end storage of the collection relevant history in EngageAR in a touch-less way. To increase automation, My Inbox also features an email scheduler component that AR professionals can leverage to send automated payment requests and invoices delivery to the clients. Furthermore, the inbox also houses a letter template functionality which allows users to create custom letter templates that simplify, and speed-up email communications being sent. The My Inbox module was deployed globally for all IBM AR professionals who covered approximately 1M invoices since January 2023.
With AI as the forefront of IBM’s vision, the EngageAR Inbox has also introduced EVA (EngageAR Virtual Assistant). EVA uses a variety of Large Language Models (such as WatsonNLU and others) to analyze client responses to provide collectors with a recommended best action such as payment forecast, dispute initiation, or statement requests. This recommendation system relies on two fundamental components: explain ability algorithms that analyze focus words within the emails and NLU which uses these focus words and classification labels to provide justification for the recommended actions. The model allowed an accuracy of 90%. EVA’s recommendation provides a brief explanation of next best action, aiding collectors in their decision-making process.
The EngageAR Inbox and Virtual Assistant (EVA) use new creative approaches to classify the received emails and provide a reason for each classification. The solution is built as an API and well suited to be re-used for other similar need.
Using past emails received from clients, a model has been created to perform a classification on those emails. With a focus on improving the accuracy, an even better solution was developed in support of EVA: combining 4 different models as an ensemble allowed drastically improve the quality of classification, achieving 90% accuracy.
- If you are nominating a brand-new product or service, state the date on which it was released. If you are nominating a new version of an existing product or service, state the date on which the update was released:
My AR Inbox was officially launched in January 2023, followed by the EngageAR Virtual Assistant (EVA) deployment in July, 2023.
- Outline the market performance, critical reception, and customer satisfaction with the product or service to date. State monetary or unit sales figures to date, if possible, and how they compare to expectations or past performance. Provide links to laudatory product or service reviews. Include some customer testimonials, if applicable (up to 350 words):
Total 334 words used.
Eliminating the use of multiple email platforms, and automating collection has assisted with simplification of sales and AR execution. The inclusion of AI functionality, which was introduced with EVA, has added a new element to IBM’s account receivable process which is a first of its time.
- Enhanced AR execution experience: Collectors can now work directly within EngageAR for collection rather than use multiple applications for client email communications. Along with inbox functionalities (Email Scheduler, Email template) it has improved productivity hours by 10K.
- Proactive Insights: With Inbox and EVA deployment collectors have received over 96% of recommendations for a next best action based on the AI-driven analysis of the content from client emails.
- Collection Remark Automation: 100% reduction Invoice collection remarking from client email responses (from 200K emails processed and over 80K client responses)
- Automated Collection Recommendation: 96% of Auto-Recommended Action provided from all client emails received (30K).
- Operational Savings: Operational savings of 164K for 2023.
- Productivity Gains: Over 10 thousand hours of collection productivity has been saved with this deployment.
These volumes demonstrate the project's remarkable success in enhancing productivity and efficiency.
We took the benefit of Large Language Model to improve the model (such as NLU) accuracy. Several other models (BERT variations) were tested to select the most accurate one, however using four different models in parallel, with a logistic regression applied to its output to select the best one resulted in a 90% of accuracy.
Each classification recommendation returned to user is also explained. Using BERT model approach, a brief sentence explains to user how the classification was done. It gives examples of email content which drove to that classification. The user has the capability to easily assess how the tool did the recommendation and provide feedback which is captured for further model improvements!
“ It is so much easier to work with the AR Inbox, having all the data required at your fingertips saves time and allows me to be more efficient “ – Adam Gonzalez, Canada AR Operations Professional
| Attachments/Videos/Links: |
|---|
|
