Skip to main content
Documentation
Documentation
Administrator Guide
Last Updated: 2026-02-06
Steps: Set Up Machine Learning for Pay Anomalies

Steps: Set Up Machine Learning for Pay Anomalies

You might need to take additional steps to enable this feature based on your organization's subscription service agreement. Your organization is either on the Main Service Agreement (MSA) or the Universal Main Service Agreement (UMSA). To determine your organization's subscription service agreement:
  1. Select your profile avatar on Workday Community.
  2. Select
    Profile
    .
  3. On your profile page, select your organization's name, which is beneath your name and next to your job title.
  4. View your
    Subscription Service Agreement
     value.
If the value is:
  • UMSA
    , the feature is automatically available. You can skip the
    Enable Innovation Services Feature and AI Data Contributions for MSA Customers
     step. For more information on Machine Learning data contributions, see Concept: Workday AI for Universal Main Subscription Agreement Customers.
  • MSA
    , you must enable this feature through Innovation Services using the
    Enable Innovation Services Feature and AI Data Contributions for MSA Customers
    step.
Note: UMSA customers don't have Innovation Services tasks and reports in their tenants as these are for MSA customers only. UMSA customers can ignore all information regarding Innovation Services.
The instructions in this topic apply to:
  • Payroll for Australia
  • Payroll for Canada
  • Payroll for the UK
  • Payroll for the U.S.
Workday Pay Anomalies uses machine learning to analyze and identify suspected payroll result abnormalities on an automatic and continuous basis. This can help improve the accuracy and efficiency of your payroll audit process. See Workday Pay Anomalies FAQ Demo for a demonstration on how you can use this feature.
The
Payroll Machine Learning GA Features
 innovation services and the
Pay Anomalies
 report are available for Payroll for the U.S., UK, Canada, and Australia but isn't functional for Payroll for France customers.
  1. On the
    Available Services
     tab, select
    Payroll
     category and then the
    Payroll Machine Learning GA Features
     check box.
    Security:
    Manage: Innovation Services domain in the Innovation Services
     functional area.
    You must opt in to all these 6 data categories to build tenanted models and to enable Pay Anomalies functionality:
    • Anomalous Payroll Results
    • Payroll Data
    • Payroll Data (Historical)
    • Pay Anomalies Validations
    • Previous States of Payroll Results (Historical)
    • Worker Data
    Not selecting all data categories will result in failing to build and deploy the machine learning model as designed. The feature won't be fully functional without these models. See Reference: Machine Learning Data Contributions.
    You might need to take additional steps to enable this feature depending on your organization's subscription service agreement. For more information, see this Community article.