Development and Retention Metrics: Getting Started

Assessing retention and development (how do you do it, how much do you assess, where do you begin, how much of the organization do you include, and so on) can be particularly difficult. The questions are endless and the temptation arises to make a declaration that development, of course, is good for people, so it must be good for the organization. However tempting and rational this argument may seem, HR simply must have more accountability today. And, while the idea that development is “good for people” is not enough, the process of assessment cannot be so overwhelming that it often gets pushed aside.

As the preceding discussion illustrates, a variety of metrics are being offered today. As we set out to review existing metrics and draft a useful set of new metrics, we used these key criteria:

• Parsimonious: Limiting the number of key criteria instead of using a lengthy scorecard.

• Usable by all companies: While we endorse creating company-specific metrics and scorecards, for the purposes of this book, the metrics should be applicable to all companies.

• Provide leading and lag indicators: Leading indicators predict future performance while lag indicators report on past performance. Leading indicators are especially important because they provide management with early warning signs of future problems.

• Balance new data with existing data: While it’s easy to create metrics that require entirely new information systems, it’s not practical. The most effective method is to create metrics that incorporate some existing and easily accessible data with new data.

• Useful to management: As seen from the HRD approach to measurement, metrics that are not useful to management are not likely to be completed.

• Correlated with performance: The metrics should have a reasonable chance of predicting performance.

• Not easily manipulated: The metrics need to be credible and not easily manipulated.

• Based on systematic information: Too often, development metrics are based on anecdotal information. The best metrics provide hard data.

• Minimally impacted by extraneous factors: The metric must reasonably represent organizational factors without incorporating unrelated extraneous factors.