Webinar - How to Implement a Data-Driven Compensation Strategy

Join Payscale’s Chief Product Evangelist, Ruth Thomas; Director of Product Management - Data Products, Vicky Peakman; and Director of Data Science, Sara Hillenmeyer as they explore the intricacies of how to design and implement your compensation data strategy.

1. How to Implement a Data-Driven
Compensation Strategy

2. Today's Presenters
Ruth Thomas
Chief Evangelist
Vicky Peakman
Director of Product, Data
Sara Hillenmeyer, PhD
Senior Director, Data
Science

3. Today's Agenda
•How to build the best data strategy
for your organization
•Other considerations:
•Data transparency
•Data bias
•Data and AI
•Q&A

4. A compensation data strategy has three components
CommunicationInput Data Internal Process

5. How transparent will we
be?
What will we provide to
employees?
How will we explain the
data sources?
What more do we need to
know?
Communication
Coverage
Repeatability
Methodology and
Explainability
Freshness
Biases
Input Data
Aggregation
Integrate internal data
Comparable jobs
Ranges
Job Postings
Internal Process
A compensation data strategy has three components

6. How transparent will we
be?
What will we provide to
employees?
How will we explain the
data sources?
What more do we need to
know?
Communication
Coverage
Repeatability
Methodology and
Explainability
Freshness
Biases
Input Data
Aggregation
Integrate internal data
Comparable jobs
Ranges
Job Postings
Internal Process
A compensation data strategy has three components

7. How transparent will we
be?
What will we provide to
employees?
How will we explain the
data sources?
What more do we need to
know?
Communication
Coverage
Repeatability
Methodology and
Explainability
Freshness
Biases
Input Data
Aggregation
Integrate internal data
Comparable jobs
Ranges
Job Postings
Internal Process
A compensation data strategy has three components

8. Data
transparency

9. Data Transparency
•Pay transparency is on the rise.
•The primary goal of pay transparency is to
promote fairness, equity, and trust by
ensuring that employees understand how
their pay is determined and how it
compares to that of their peers.
•In practice this means openly sharing
information about employee compensation-
including how it is determined.
•This can extend to transparency of data,
including sources and data strategy.

10. The pay transparency continuum
Pay
Secrecy
Pay
Transparency
No information
shared with
employees
except their
own pay
Share salary
range data for
current role with
employee
All salary range
data shared
internally
Salaries
shared
publicly
Individual
employee
salaries shared
internally
Share with
individuals how
their pay is
determined

11. The pay transparency continuum – data strategy
Pay
Secrecy
Pay
Transparency
No information
on data strategy
shared with
employees or
managers
Share with
employees pay
philosophy
Share data
strategy with
managers that
is used to
create range
All data is
shared with
managers
Share data
strategy with
employees that
is used to
create range
Share with
managers pay
philosophy
All data is
shared with
employees

12. Poll 1: How
transparent are
you with your
data strategy?
•We communicate to HR or senior management only
•We communicate this to managers only
•We communicate this to all
•We want to move towards transparency
•We arenot transparent currently
•I don't know

13. Data bias

14. All market data is biased
•Surveys/Peer are collected from companies that participate (generally skews toward larger
companies)
•ERD and other employee-reported data (glassdoor, levels.fyi, etc) often skews towards
people who look up their pay information online (younger, more technical)
•Aggregated data from job postings are biased towards states that have pay transparency
legislation
•Bureau of Labor Statistics tries to “even out” their data to get make it a representative
sample of labor in the US. Most compensation data providers do not do this! (None, that I
know of.)

15. Peer, like many HR-participation-based data sources, is
biased toward larger companies
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00
80.00
90.00
100.00
<5 5-9 10-19 20-49 50-99 100-249 250-499 500-999 1000
% of Total Incumbents in Dataset
Number of Employees at Company
% Incumbents working for a company with X Employees
Bureau of Labor Statistics ERD Peer

16. Peer is biased toward Retail companies;
ERD toward Professional, Scientific, and Technical Services
0.00%
5.00%
10.00%
15.00%
20.00%
25.00%
30.00%
35.00%
Health care and social
assistance
Retail trade
Accommodation and food
services
Manufacturing
Professional, scientific, and
technical services
Construction
Finance and insurance
Transportation and
warehousing
Information
Arts, entertainment, and
recreation
Real estate and rental and
leasing
Utilities
Wholesale Trade
% Incumbents in Industry Group
Data Product Industry Bias versus Bureau of Labor
Statistics
Bureau of Labor StatisticsERD Peer

17. What should you do? It depends!
•You may WANT biased data! (if it
matches your pay market)
•Sometimes the bias doesn’t matter
•Sometimes you can counter-act the
overall bias by sub-setting the data
(using a scope)
•Sometimes you may want to counter-
act the bias with an adjustment factor

18. Where can bias creep into the process?
Inconsistency
in eligibility for
Incentive Plans
Matching /
pricing one
by one
Inconsistency in
survey use
Differences in
treatment of Part-
Time vs Full-Time
roles
Ad-hoc pay
adjustments
Inconsistency in
market reference
points / weightings
/ aging factors
Differences in
applying differentials
(eg skills, education,
location)
Lack of
understanding
of comparable
jobs
Allowing manager
discretion
Use of
performance in
setting pay
Different people
matching / pricing
Matching /
pricing using
salary as a guide
Creating
structures using
salary as a guide

19. Poll 2: Do you
monitor your
data processes
for bias?
•Yes, continuously
•Yes, occasionally
•No, but we are planning to
•No, I was not aware of the issue

•I don't know

20. AI as part of a
data strategy

21. AI sentiment
Payscale CBPR 2024

22. Payscale CBPR 2024

23. AI as part of data strategy
Payscale CBPR 2024
0%
10%
20%
30%
40%
50%
60%
Undecided Cautiously
Optimistic
Against It Totally on board
What is your organization’s overall
sentiment around using AI in making
compensation decisions?
Forms of AI are already widely used:
•Aggregating data
•Aging data
•Calculating and applying
differentials for location, industry,
company size
Future:
•Improved precision/accuracy in
calculated market ranges
•Better de-biasing of underlying
data sources

24. Now and near-term use cases for AI in compensation
Survey
matching
automated
Benchmark
jobs and predict
pay ranges
Salary offer
acceptance and
ROI analysis
Geo & skills-
based pay
adjustments
recommended
Comp trends &
market monitoring
with alerts
Pay increase
recommendations
and pay equity
monitoring
Job descriptions
auto-generated
Total Rewards
Statement (TRS)
auto-generated
Pay communications
auto-generated

25. Things AI Cannot Do…
Eliminate bias
in pay
Eliminate
human review
and oversight
Determinecomp
strategy
to achieve key goals

26. Poll 3: How are
you using AI
(select all that
apply)?
•Use AI to benchmark and price jobs or predict pay ranges
•Use AI to monitor for pay equity and/or suggest pay increases
•Use AI to collect intelligence on skills for recruiting,
education/upskilling, career pathing, and/or compensation
•Use AI to write offer letters and generate total rewards
statements
•Use AI to speed up survey participation and year-over-year
updates
•None of the above
•I don't know

27. Recommendations
Aim for consistency of methodology
Aim for transparency so you can explain each number
Understand the biases in the data and in your processes
Understand how AI can help you now and in the future

28. Intelligent streams of curated, validated, compensation data
Payscale’s Diverse & Dynamic Data Portfolio
Peer
A transparent & dynamic
HR reported data network
100 M salary profiles (all time)
40M salary profiles in use
350,000 new profiles/month
15,000 jobs
8,000 skills/certifications
1 billion+ data points
4,900 jobs
15 countries
2,400 organizations
7M employees
10,000 surveys
From 300+ publishers
Employee Reported
The world’s largest
real-time salary database
HR Market Analysis
A composite of analyst curated
employer reported survey data
Published Survey Data
Trusted data partner
4,500 jobs
100+ industries
Compensation Survey
A modern, quarterly
compensation survey
1,350 organizations
2.9M employees
6,111 jobs
Request a demo of
Payscale data in the polls
tab now!

29. Q&A
Feel free to ask any questions in the chat!