Exploring Now Assist for Accounts Payable Operations (APO)
Summarize
Summary of Exploring Now Assist for Accounts Payable Operations (APO)
The Now Assist for Accounts Payable Operations (APO) application leverages generative AI to streamline the accounts payable process. It enables AP fulfillers to summarize cases, auto-map invoice lines with purchase orders, and extract invoice details efficiently, enhancing case management and processing accuracy.
Show less
Key Features
- Invoice Line Mapping: Automatically maps invoice lines to purchase order lines, facilitating a review process for accuracy before finalization.
- Invoice Case Summarization: Generates concise summaries of invoice cases, which can be updated in real-time as case details evolve.
- Invoice Data Extraction: Extracts details from invoices submitted via email or supplier portals in multiple languages, improving accuracy and reducing manual work.
- Invoice Inquiry Solution Generator: Automates resolution suggestions for invoice inquiries by analyzing case details and related records.
- Relevant Invoice Finder: Utilizes semantic matching to identify historically processed invoices relevant to current inquiries.
- Email Response Generation: Drafts email responses based on invoice case context, improving response time and productivity.
- Sentiment Analysis: Prioritizes inquiries based on the emotional tone of the responses, helping fulfillers address urgent cases effectively.
- Document Classification: Automatically categorizes supplier documents (invoices, credit memos) for streamlined processing, ensuring non-invoice documents are not misclassified.
Key Outcomes
By implementing Now Assist for APO, ServiceNow customers can expect a significant reduction in manual effort, improved accuracy in invoice processing, and enhanced productivity for AP specialists. The automation of case summaries and email responses allows fulfillers to focus on more complex tasks while ensuring timely and efficient handling of accounts payable operations.
With the Now Assist for Accounts Payable Operations (APO) application, an account payable (AP) fulfiller can use generative AI to summarize the details of a case.
Now Assist for APO overview
Using the summarization capability of the Now Assist for APO application, an AP fulfiller can auto-map the invoice lines with purchase order lines, get a concise summary of a case quickly, extract invoice details from email and auto-create an invoice processing case.
Cases can be complex and a case summary may enable an AP fulfiller to provide a more concise summary.
Skills
- The generative AI skills enable an AP fulfiller to extract the details of an invoice case or inquiry case and suggest resolutions.
- The generative AI skills perform the purchase order line-mapping and enable an AP fulfiller to review the mapping results.
- Purchase order line mapping
- Provides AP fulfiller to automatically map the invoice lines with the purchase order lines.
The purchase order line-mapping skill generates the invoice line mapping, changes the invoice line status to review mapping state and provokes the fulfiller to manually review the mapping results. After the fulfiller reviews the mapping results, the invoice line status is changed to mapping complete. The purchase order lines mapped using the Now Assist skill are controlled using the APO properties. For more information on the use of purchase order mapping in workspace, see Set APO properties and Using Playbook in Accounts Payable Operations.
- Invoice case summarization
- The summary skill reviews the issue, actions completed, status, and next steps. An AP fulfiller can refresh the summary as case updates happen and post the summary to work notes.
The invoice case summary displays above the activity stream.
For invoice processing cases, the summary skill also reviews the invoice status and invoice information located in the applicable document.
For inquiry cases, the invoice case summarization skill generates a case summary by checking the short description, description, and tasks that were created for the case.
Fulfiller can customize prompt configuration and prompt optimization using the pre-processor in the Now Assist skill kit (NASK). The skill supports multiple models such as OpenAI, Claude, Gemini, Now LLM Service. The configured pre-processor is applied across all supported models in NASK. To customize the prompt instructions, the sn_skill_builder_admin role is mandatory.
- Invoice data extraction
- The Invoice data extraction skill helps in extracting invoice details from PDFs submitted through email or supplier portal. The digitized invoice is processed within Accounts Payable Operations automatically. Invoices are extracted in languages (English, Dutch, French, German, Japanese, Italian, Portuguese, Spanish) supported by GPT-4o.
With the Invoice data extraction skill, customers can achieve higher invoice extraction accuracy by leveraging GPT-4o and significantly reducing manual effort. Customers are provided with the flexibility of turning on the manual review of extraction results or can let invoice be processed through the workflow automatically without manual review.
- Invoice inquiry solution generator
- The Invoice inquiry solution generator skill identifies, extracts the invoice inquiry details from the invoice data and related records. Automates the resolution for invoice inquiry cases.
- Relevant invoice finder
- The Relevant invoice finder skill identifies the historically processed invoices that are most relevant to a given invoice using semantic matching.
The skill compares the current invoice and its line descriptions with previously processed invoices along with its line descriptions, and generates a relevance score from 0 through 1 for each item. These invoices are then returned in a ranked list, ordered by relevance.
The Recommend invoice owner AI agent uses this match score to suggest a resolution plan for Missing or invalid business owner exceptions.
- Email response for invoice case
- The Email response for invoice case skill generates responses for email by analyzing invoice case contexts, activity, work notes, invoice details, approvals, payments, fulfiller's response in the past and knowledge base
articles. The skill enables AP specialists and agents to automatically draft and manage responses to invoice-related inquiries. It also leverages the templates in drafting email responses.
The skill reduces time spent by AP specialists and agents in manually drafting responses and improves productivity.
- Email response for invoice task
- The Email response for invoice task skill generates responses for email by analyzing invoice case contexts, activity, work notes, invoice details, approvals, payments, fulfiller's response in the past and knowledge base
articles. The skill enables AP specialists and agents to automatically draft and manage responses to invoice-related inquiries. It also leverages the templates in drafting email responses.
The skill reduces time spent by AP specialists and agents in manually drafting responses and improves productivity.
- Sentiment analysis for invoice case
- The Sentiment analysis for invoice inquiry case skill enables fulfiller to analyze and prioritize the invoice inquiry cases based on the response tone. The tone is categorized as positive, negative, or neutral. The sentiment score helps fulfiller to understand the urgency and emotional weight of the invoice inquiry case.
- Accounts payable document classification
- The Accounts payable document classification skill automatically identifies and categorizes supplier sent email attachments as invoice, credit memo, and supporting documents. The system supports configurable confidence thresholds and an AI solution with confidence score. Documents are auto-approved if the AI-confidence score is greater than the configured threshold, and auto-approval is enabled in the APO properties page. AP specialist manually reviews the document if the AI confidence score is lesser than the configured threshold. After the documents are successfully categorized, the invoice data extraction skill triggers the extraction of invoices.