Machine Learning for the Frontline
Workforce: Where We Are & Possibilities

Machine Learning for the Frontline Workforce:
Where We Are & Possibilities
Looking Forward with Workday is a webinar series designed to give you insight
into how your organization can do more with Workday. This series will share how we
support how you manage your people and your business. Topics will include solution
strategies and roadmaps, specific use cases, live demos, and customer stories.

This presentation may contain forward-looking statements for which there are risks, uncertainties, and
assumptions. If the risks materialize or assumptions prove incorrect, Workday’s business results and directions
could differ materially from results implied by the forward-looking statements. Forward-looking statements include
any statements regarding strategies or plans for future operations; any statements concerning new features,
enhancements or upgrades to our existing applications or plans for future applications; and any statements of
belief. Further information on risks that could affect Workday’s results is included in our filings with the Securities
and Exchange Commission which are available on the Workday investor relations
webpage: www.workday.com/company/investor_relations.php
Workday assumes no obligation for and does not intend to update any forward-looking statements. Any
unreleased services, features, functionality or enhancements referenced in any Workday document, roadmap,
blog, our website, press release or public statement that are not currently available are subject to change at
Workday’s discretion and may not be delivered as planned or at all.
Customers who purchase Workday, Inc. services should make their purchase decisions upon services, features,
and functions that are currently available.
Safe Harbor

Help Slides Media
Player
Resource
List
Q&A
Housekeeping Items
Speakers Survey
Housekeeping Items

Jackson Bryant-Comstock
Sr Manager, Product Management
Workday Workforce and Pay
Goli Cobbe
Manager, Product Management
Workday Machine Learning

Agenda
The Frontline Workforce
Workday’s Focus
Machine Learning
Roadmap
Q & A

Workday’s Focus on the Frontline
Covid-19 has reminded us more than ever how
essential the workers on the frontlines are. They also
represent a major subset of the global workforce
- In the US Today, over 55% of all workers are
hourly
- Gig, Contract and Hourly worker numbers are all
on the rise across the globe
- Focus on industries that dominate this worker
persona including retail, food services, healthcare
and more

Frontline Centric Solutions
Geofencing Mobile
Check In
Pay On Demand
Scheduled vs Actuals
Detail
Worktag Automation

Workday for the Frontline Workforce
Expand Functional Capabilities Create an ideal user experience for
mobile workers and managers
Expose Business Value through
delivering prescribed analytics
Drive efficiencies through targeted
machine learning

Workforce and Pay ML Opportunities
PayrollTime Tracking
Absence
Management
Core HR
Mobile

Workforce
Management is
Ripe for ML
Disruption
Data, Data and
more Data
10+ Million Active Time &
Absence Users
Frequent
Feedback Loop
Weekly business process
approvals, corrections
and entry allow for fast
ML training
Fast Business
Value
Time Saved is Time
Back for your most
valuable workers

Opportunities for Machine Learning
Application
Time Entry Time Approvals Pay
Worker, job and talent demographic data provide contextual insight to make increasingly more
accurate ML driven suggestions to global organizations

Success for Our Customers
Deliver differentiated value by giving your most valuable
users time back in their day to do what matters most

Machine
Learning for
the Frontline
Workforce

Machine Learning Frontline Worker Focus Areas
• User Experience
• Business Value and Cost Savings
• Low-touch, effective, automated
• Helping create an engaged and productive workforce
• Simple and personalized

Building Intelligent Computing Into the Core
of Our Products
Prism
Analytics
Workday
Human Capital
Management
Workday
Financial
Management
Workday Planning

Early Time Entry Error Detection Has Downstream
Benefits
Worker: Submits
Time
1
Payroll Admin:
Approves or Denies
Errors can result from Time Issues
3
Manager: Approves
or Denies
Catching errors at this stage
prevents a future Payroll error +
High Administrative costs
2
4
Worker Is Paid
Workday confidential

Complete ISA and
opt in
Models are deployed in
tenant & continuously
updated (off cycle)
Models are
Trained
Machine Learning EA Customer Engagement

Planning &
Discovery
Identify high impact
use cases
Data
Extraction
Gain access to
customer data via
ISA
Development
Explore data. Select
Algorithms. Build
model pipeline
Deployment
Release to EA
customers
Iterate
Use customer
feedback and data to
continuously
improve ML model
More feedback. Better predictions.
Virtuous cycles
Our ML Dev Approach

Machine
Learning Dev
How Do We Chose the Right Model?
ML Journey
Plan
Train
Validate
Scalable High Quality

Time
Anomalies

Frontline Manager “I review my worker’s time cards for a week,
bi-weekly period or month. This is a tedious
task that I’d rather spend as little time on as
possible, but it’s my responsibility to make
sure that time is accurate for payroll, project
invoicing and audits. I’d ideally like
anomalies surfaced to me so I can make the
best use of my time.”
Matt
Frontline Manager
● Costly mistakes
● Spending too much time on manual
review
FEARS

Time Admins Spend:
On Administrative Work
2 days/week
8 days/month
Reaching out to workers who forgot to report time
Analyzing time for exceptions
Reaching out to managers who forgot to approve
Reaching out to workers who forgot to submit time
Make individual corrections as needed

Frontline Manager “As a manager, I want to be alerted when my team’s
specific time entry codes, projects, tasks, worktags or
quantity of time entered is anomalous compared to
what is expected based on past entries.”

Goal: Maximize True Positives and Negatives
True Positive
Time to action well spent
False Negative
Completely missed anomaly
False Positive
Not a good use of time
True Negative
No action required
ACTUAL
NO ANOMALYANOMALY
PREDICTED
“I think there is
ANOMALY”
(+)
“I think there is
NO ANOMALY”
(-)
🙂
🙂

Vision
Manual, labor intensive, and
error prone
Deadlines are already tight
Inexhaustive: can’t review every associate
Identifies trends not visible to a single
individual
Helps timekeepers prioritize items
needing attention
Improves manager confidence and time
back in their day
0 5 10 15 20 25 30 35 40 45
Calendar Day
Traditional Pay Period Close
Analysis
Close Activities
Business Activities
0 5 10 15 20 25 30 35 40 45
Calendar Day
Analysis
Close Activities
Business Activities
Machine Learning Infused Close
Workday Confidential

Feedback from
Customers
Wow, this will save me
time
This is useful
We don’t populate that
field
We use a 3rd party
system to store that
information
We added a new
location in the UK

W & P - Time Anomalies on Review Time
• Filter time entries by different categories including
Time Anomalies
• Workers with Time Anomalies are sorted by count
Reduces the amount of manual review required in
reviewing timesheets, frees up time for other tasks.
What
Benefit / Value

The Tip of The
Iceberg

Payroll Insights

Persona: Payroll Analyst
““Meet deadlines, no errors.
That’s what it is.”
Pat
● Getting delayed due to my
inability
● Not being able to explain
reasoning behind my payroll
processing
● Many service tickets from
employees (“customers”)
questioning paycheck results
FEARS
● Process payroll in time and
successfully without errors
● Store all payroll trails to be
prepared for any future audits
(by leads, employee, external auditors)
GOALS

Time and People Value
Time Value:
➔ Organizations spend up to 80% of pay cycle on reviewing payroll
➔ 50-80% of payroll review is spent on just identifying anomalies
People and Business Value:
➔ Opportunity to save time for payroll teams, allowing efforts
to be spent on more strategic and higher priority tasks
➔ Opportunity to narrow the pay tolerance interval,
decreasing costs and increasing pay accuracy.

Settlement/ACHPay Input
Integration
Pay Calc Pay Complete
File from External
Vendor
(Time, Stock,
Benefits..)
Pay Inputs
Corrections
Payroll Insights
Pay Professionals
How Payroll Insights Can Help

Assistant
• Simple navigation, task
execution, and information
retrieval through natural
conversation
• Start a conversation
• Available in Workday Desktop,
Workday Mobile, Microsoft
Teams, and Slack
Workday Assistant

• Time Anomalies
• Payroll Insights
• Additional Workday
Assistant tasks
At a Glance: ML for Workforce and Pay
Workday Confidential
Model Training & Validation Early Access Release General Availability
• Customer partnership
• Model training
• Qualitative and
Quantitative
measurements
• Time Anomalies in EA
• Payroll Insights in EA
FUTURE

Key Takeaways
Workday Confidential
Investments at the core have
acted as a flywheel to
accelerate delivery of ML
capabilities over the past
year.
Tip of the Iceberg
This year was the tip of the
iceberg. There’s even more
coming from Workday to
make better decision makers
and more efficient workers.
Embedding Machine
Learning into the very core of
Workday enables us to more
efficiently provide ML
capabilities that deliver real
value.
Flywheel Effect
ML Foundation

Machine Learning at Workday:
https://community.workday.com/machinelearning
Innovation Services: https://community.workday.com/articles/409032
ML Roadmap: https://community.workday.com/articles/698718
Access More
Resources
Workday Confidential

Q&A

Thank you for joining us!
Join us in these upcoming sessions:
Workday Confidential
December 1, 2020
Tuesday 10 AM PT
Workday on Workday: Best
Practices for Building a Financial
Data Hub with Workday
Financial Management and
Prism Analytics
December 8, 2020
Tuesday 10 AM PT
Belonging & Diversity: Learn
how Netflix is using Workday’s
Solutions to VIBE in their
Workplace
December 15, 2020
Tuesday 10 AM PT
Why Investing in Digital
Technologies Now Can Radically
Change Your Business
December 16, 2020
Wednesday 10 AM PT
Using Prism to Manage Your
Data Through an M&A: Learn
How Warner Bros is Using
Workday’s Prism Analytics to
Establish One Source of Truth
December 17, 2020
Thursday 9 AM PT
Unlock New Value from your
Workday Investment: Discover
the Power of Workday Extend
Apps

For a changing world.
TM