Applying Analytics Can Enhance Workforce Planning

Turning “people data” into an action plan for talent management is a strategic opportunity for HR professionals.

Management expects HR professionals “to be corporate soothsayers” regarding talent needs, “but our predictions should be based on an analysis of data, not just hunches,” said Kelly Rene Wenzel, SHRM-SCP, vice president for analytics for the OptFocus product suite at OptTek Systems Inc., a Boulder, Colo.-based business software and services firm.

“We’re the custodian of workforce information, but we’re not making full use of it,” she said, speaking at a preconference workshop at the Society for Human Resource Management 2016 Annual Conference & Exposition on June 18.

Workforce analytics can reduce workforce-related costs and risk, added Marco Better, vice president for business development at OptTek. Through computer-based simulation modeling, “you can explore alternative recruiting and staffing scenarios. This ensures you have the right kinds of talent and the optimum number of individuals to achieve your goals with predictive accuracy,” he noted.

Analytics can reduce workforce-related costs by eliminating much of the guesswork that puts organizations at risk for overstaffing, understaffing and other costly workforce problems, the presenters said.

“ ‘Big data’ is an evolving term that describes any voluminous amount of data that has the potential to be mined for actionable information,” Better said. This data helps HR people to answer top management when they’re asked, “What happens when …?”

Avoiding Talent Gaps

Determining “the talent we have versus the talent we need—or soon will need” is a key HR challenge. But HR’s response is often reactionary, such as “discovering a talent gap after people quit,” Wenzel said. “Some global corporations can’t even tell you what their head count is,” Better added.

When organizations have more people than they need, the result is unnecessary costs that make management unhappy. Conversely, when the talent supply is stretched too thin, workers are unhappy and more likely to leave, creating added recruitment and replacement costs—not to mention disengagement issues among those who don’t leave.

Applying analytics to talent supply challenges involves looking at the history of attrition—that is, why people are leaving. This also involves determining whether it costs more to retain employees or to let them leave and then hire from outside, which may depend on the particular job in question, and the ease or difficulty of filling that position.

“If there are issues with attrition and recruitment is falling short, focus on improving retention, increasing training and promotions, enriching the perceived value of benefits, making improvements in the culture,” Wenzel advised.

One response may be to improve benefits, such as by adding in-house day care. But rather than just guessing, “Use predictive analytics to make that determination” of what benefits to add, Wenzel said. “Is it a draw for employees? Are there analytics to back that up?”

Which brings up “optimization”—determining the best investment level for optimal returns. “Putting more money into benefit programs doesn’t necessarily increase returns” by raising retention rates, Wenzel observed.

“Through historic analytics, we can pattern data to make predictions,” Better said. “To determine if high-value employees are likely to stay or go—and why—determine the retention pattern among employees with similar characteristics.”

Simulate the paths of employees as they go through their careers, for instance, by incorporating avatars into the model, Better suggested.

Modeling with Workforce Metrics

The presenters suggested various metrics to use in simulation modeling and predictive analytics.

Descriptive past:

• Historical retention rates.

• Historical transfers and promotions.

• Historical training.

• Historical performance metrics.

• Historical costs and budgets.

Proactive future:

• Impact of potential plans and actions on the talent supply.

• Impact on workforce quality and engagement.

“Model the workforce at a granular level, down to role levels,” Better advised. “Use the model to predict future behavior and test the effect of changes to the workforce. Not every gap is the same, so create plans for specific roles.”

The model should describe the behavior of the workforce by including data related to:

• Retention (voluntary, involuntary, retirement).

• Mobility (promotions, transfers).

• New hires and sourcing.

• Time to proficiency.

“The model can be extended to incorporate new factors and variables as they become available, such as social networking data and external data such as unemployment levels,” Better said.

When using modeling for talent supply/demand gap-risk analysis, include metrics for:

• Gap risk by role/profile.

• Talent flow by role and by period.

• Head count growth.

When conducting attrition-risk analysis, look at data for:

• Voluntary attrition, involuntary attrition, retirements.

• Attrition risk by role and demographics, including new employees, high-performers and diverse populations.

For talent-acquisition analysis, consider:

• Hiring rates (internal and external).

• Sourcing channel analysis.

• Hiring costs.

• Contingent vs. full-time employees.

• Quality of hiring practices.

Telling a Story

When communicating your findings to management, “Use the data to tell a story,” Wenzel said. “Don’t show them the spreadsheets—that just bores them, and they’ll interpret the numbers the way they want to.”

Focus on engaging the people you need to persuade, she advised. When meeting with management, “Go in with a business need to solve,” such as fixing the talent gap, “then show them how your findings lead to improved cost-savings and revenue enhancement.”

Stephen Miller, CEBS, is an online editor/manager for SHRM. Follow me on Twitter.

Related Resource:

Workforce Analytics Case Study, OptTek Systems Inc. 

Related SHRM Article:

Report: Make Workforce Analytics Work for Business, SHRM HR News, June 2016