Human Capital Analytics: Introduction Notes

Introduction: Realizing the Dream: From Nuisance to Necessity

Main Idea

The introduction explains how HR and training evolved from being viewed as low-value support functions into areas that are increasingly expected to show measurable business value through human capital analytics.

The author’s main argument is that organizations depend heavily on people, but HR historically struggled to prove its value with data. Analytics gives HR a way to connect people-related decisions to organizational outcomes.


Starting From the Back Row

When Jac Fitz-enz joined Wells Fargo in 1969, Personnel and Training were located in an annex away from the main building.

This reflected how HR was viewed at the time:

  • Peripheral to the business

  • Not strategic

  • More focused on employee happiness than measurable organizational contribution

Bankers focused on numbers like:

  • Revenue

  • Profit

  • Costs

Personnel and Training mostly focused on people and satisfaction without connecting those efforts to business results.

Key takeaway

The issue was not that employee satisfaction was unimportant. The issue was that HR could not clearly demonstrate:

How its work created value for the organization.


The Value Dream

Beginning in the 1950s–1970s, several people began arguing that HR and training should be evaluated more systematically.

Important contributors mentioned include:

  • Don Kirkpatrick — developed the foundation for the Four Levels of Evaluation.

  • Ray Killian — wrote about ROI in HR.

  • Jack Phillips — became a major contributor to training evaluation and ROI.

  • Jac Fitz-enz — focused heavily on HR measurement and metrics.

Fitz-enz later founded the Saratoga Institute, which developed HR metrics and benchmarking systems.

The original goal was partly defensive:

Show management that HR was not just a cost center.

HR could contribute to both:

  • Revenue growth

  • Expense reduction


Barriers to HR Measurement

Many HR professionals initially resisted analytics.

Common beliefs included:

  • People should not be reduced to numbers.

  • Measuring employees could be dehumanizing.

  • Human value could not be quantified.

The author argues that avoiding measurement creates another problem: traditional accounting already treats employees primarily as expenses, not assets.

Therefore, HR needs evidence to show that investments in people create organizational value.

Good discussion point

There is a tension between:

Measuring people to demonstrate their value

and

Reducing people to financial or numerical outcomes.


Organizations Are About People

The author argues that people are the only truly active organizational resource.

Organizations may own:

  • Technology

  • Equipment

  • Money

  • Buildings

  • Inventory

But these resources cannot create value without people using them.

The author therefore criticizes organizations that spend heavily on technology but fail to invest in the employees who must use it.

Cash register example

A retailer introduced new software but gave employees almost no training.

Results included:

  • Mistakes

  • Frustrated employees

  • Angry customers

  • Lost sales

The broader lesson:

Investment in technology without investment in people can reduce the value of the technology itself.


Managing Risk Through Analytics

Managers constantly make decisions under uncertainty because organizations face:

  • Competition

  • Regulation

  • Technology changes

  • Customer changes

  • Workforce issues

The author argues that experience alone is not enough because past experience may no longer fit current conditions.

Analytics provides more objective evidence for decision-making.


Three Types of Analytics

Descriptive Analytics

Answers:

What happened?

Examples:

  • What was turnover last year?

  • Which department had the most absenteeism?

Predictive Analytics

Answers:

What is likely to happen?

Examples:

  • Which employees are most likely to leave?

  • What workforce needs are likely next year?

Prescriptive Analytics

Answers:

What should we do?

Example:

If poor supervisor relationships predict turnover, what management intervention should we implement?

The author describes this approach as:

“Manage tomorrow, today.”

Easy way to remember

Descriptive = What happened?
Predictive = What will probably happen?
Prescriptive = What should we do about it?


Historic Fundamentals of HR Measurement

The author identifies five basic ways of measuring something:

  1. Cost

  2. Time

  3. Quantity

  4. Quality

  5. Human reaction

Early HR measurement focused mostly on transactions such as:

  • Cost per hire

  • Number hired

  • Number trained

  • Number retained

Eventually, HR began asking a more important question:

What effect did those activities have?

Examples:

Instead of only asking:

How many employees completed training?

HR began asking:

Did training improve productivity or reduce errors?

This represents the shift from measuring HR activity to measuring HR impact.


Benchmarking and Predictive Analytics

After organizations began measuring performance, managers asked:

How do we compare with other organizations?

That led to benchmarking.

The Saratoga Institute developed major HR benchmarks across industries and countries.

Benchmarking helped identify patterns and trends, but eventually organizations wanted to know:

Why are these patterns happening?

That helped push HR toward predictive analytics.


Intangible Assets

The introduction also discusses intangible assets — sources of future organizational value that do not have a physical form.

Examples include:

  • Human resources

  • Organizational knowledge

  • Organizational practices

The author argues that traditional accounting does not fully represent these sources of value.

This supports the larger argument that employees contribute significantly to organizational value even though that contribution may not appear clearly on a balance sheet.


The Move Toward Predictive Analytics

By the mid-2000s, HR still lacked a strong systematic approach to predicting the outcomes of investments in people.

Other functions were already using predictive analytics, including:

  • Marketing

  • Finance

  • Insurance

  • Retail

  • Healthcare

The author finds it ironic that HR argued employee behavior could not be predicted while organizations were already predicting customer behavior in much more open and complex environments.


HCM:21

Fitz-enz and others developed HCM:21 — Human Capital Metrics for the 21st Century.

One of the most important lessons from developing the model was:

Statistics should not be the starting point.

Before asking what data to analyze, organizations need to understand:

  • What problem exists.

  • What they do not know.

  • What questions need to be answered.

Only then should they determine what metrics or statistical methods are appropriate.


Practicality: Don't Start With “What Should We Measure?”

Managers often immediately ask:

“What should we measure?”

The author argues that measurement should actually come later.

Analytics should begin by understanding the situation.

A major problem with organizations is that analytics is often treated as a one-time project.

A problem appears, data are analyzed, a solution is created, and then the entire process must be recreated when the next problem appears.

Instead, organizations need a consistent and repeatable analytics framework.


Two Foundations of the Analytics Model

Human capital analytics rests on two main pillars.

Pillar 1: Logical Questions

Start by asking:

What is happening?

Identify the problem.

Why is it happening?

Identify possible causes.

What effect is it having on the organization?

Connect the human-capital issue to organizational outcomes.

What is happening inside the organization that affects performance?

Consider factors such as:

  • Culture

  • Systems

  • Processes

  • Goals

  • Strengths

  • Weaknesses

What options do we have to improve the situation?

Move from understanding the problem toward potential solutions.

A simple way to remember this:

What? → Why? → So what? → Now what?


Pillar 2: Statistics and Technology

After the right questions have been identified, organizations can use:

  • Statistical analysis

  • Technology

  • Quantitative data

  • Qualitative data

to identify solutions that are:

  • Effective

  • Efficient

  • Sustainable