Company: IBM, Armonk, New York, NY
Company Description: IBM is a values-based enterprise of individuals who create & apply technology to make the world work better. Today, more than 250,000 IBM employees around the world invent and integrate hardware, software and services to help forward-thinking enterprises, institutions and people succeed on a smarter planet.
Nomination Category: Company / Organization Categories
Nomination Sub Category: Technical Innovation of the Year - At Organizations With 1,000 or More Employees
 Nomination Title: Optimizing Cash Management with AI
- Which will you submit for your nomination in this category, a video of up to five (5) minutes in length about the achievements of the nominated organization since January 1 2022, OR written answers to the questions for this category? (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 nominated 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 providing written answers for your submission, you must provide an answer to this first question: Briefly describe the nominated organization: its history and past performance (up to 200 words):
Total 200 words used.
While the hype around AI and foundational models continues to grow, many organizations need help deploying AI tools across real-world environments, with only 54% moving from pilot to production (“Gartner 2022 AI Survey”, Gartner, 2022). Lack of a transparent process concerns AI stakeholders, where data scientists, let alone users, often struggle to trace the basis for the AI model decision.
Our real-world challenge is that to successfully manage cash flow, the client’s contractual payment terms must be consistently enforced; however, term identification is a manual and time-consuming activity open to user interpretation of contract language. The IBM Quote-to-Cash (Q2C) Transformation team deployed an AI solution to optimize IBM cash management process and drive operational efficiencies by delivering consistent insights on the client’s contractual terms with full AI model decision transparency to empower users and create a high-quality user experience.
An English-language Contracting Language Analyzer (CLA) Tool was successfully deployed in 2022 to identify terms and conditions relevant to Accounts Receivable (AR) in contracts stored in the Enterprise Contract Repository (ECR). CLA was expanded in 2023 to cover Spanish, Italian, French, German, and Portuguese, covering more than 88% of global contracts, allowing IBM to reduce aged receivables and optimize cash opportunities.
- If you are providing written answers for your submission, you must provide an answer to this second question: Outline the organization's achievements since the beginning of 2022 that you wish to bring to the judges' attention (up to 250 words):
Total 243 words used.
With implementation of CLA, the organization was able to deliver significant benefits such as:
- An effective and simplified overall AR user experience: users access via integrated tooling Enterprise Contract Repository / Contract Language Analyzer (ECR/CLA) and are presented with a simple overview of the relevant data needed.
- Significantly improved speed and accuracy to locate the appropriate contractual document, eliminating time-consuming manual effort to read entire contract to identify AR terms.
- Consistent quality of output, reducing the risk of missing or misinterpreting AR terms.
- Building foundation for future automation of specific AR processes, such as loan performance assessment for delinquent clients, to help drive excellence in cashflow management.
In addition, this initiative highlighted the volume of non-standard language across the portfolio and greater insight into understanding the cashflow impacts of negotiated terms, which reinforced the need to standardize and simplify standardized contract templates and enforce tighter adherence of non-standard terms.
The CLA tooling optimizes time and resources to drive better decisions and improve business outcomes by providing easily accessible contractual terms for cash collection:
- 88% coverage of the entire contract universe, supporting English, Spanish, Italian, French, Portuguese, and German languages via CLA.
- 100% time saving for AR professionals where AR terms are populated at invoice level.
- 86% time saving for users via Enterprise Contract Repository.
The AR terms initiative helped support the delivery of critical AR improvements YtY, representing 3Q’23 YTD global cash improvement of 5% and a reduction of aged receivables by 30% YtY.
- If you are providing written answers for your submission, you must provide an answer to this third question: Explain why the achievements you have highlighted are unique or significant. If possible compare the achievements to the performance of other players in your industry and/or to the organization's past performance (up to 250 words):
Total 249 words used.
While contract analysis tools utilizing natural language understanding (NLU) could help with contract understanding, they tend to focus more on legal than commercial analysis. Additionally, they focus mainly on class/concept classification within the contract for presentation to the user (by class label) for user review and evaluation. We could not find in the industry a system or process that can identify cash collection terms, evaluate them against standards and system of record, and extract key data such as the number of allowed payment days, % late payment fees, or whether written consent (payment assignment) is required of not.
With that in mind, we developed a method and system designed to identify the presence/absences and/or modification of key cash management contract terms within a contract, then to evaluate the identified term as standard/non-standard/missing and extract additional information such as the number of payment days, if late payment fees (LPF) are allowed and % LFP, or written consent required etc.
An unprocessed document goes through a pre-processing step where non-useful information is removed and textual content is formatted and tokenized, sentence boundaries are defined, and possible noises filtered. Then a three-step process seeks to identify and classify documents sentence as cash term classes, this is followed by a data extraction process, and finally an evaluate and warning process.
The beauty of this method, which delivers 90% accuracy, is that it was developed for cash collection use cases, but it applies and can be scaled to a wide variety of use cases.
- You have the option to answer this final question: 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):
| Attachments/Videos/Links: |
|---|
|
