The quality of human capital management is now the single most important predictor of an organisation’s business results. However, as Laurie Bassi and Daniel McMurrer write, there are a number of essential attributes that next generation human capital measurement systems possess
The quality of human capital management is now the single most important predictor of an organisation’s business results. However, as Laurie Bassi and Daniel McMurrer write, there are a number of essential attributes that next generation human capital measurement systems possess
We stand at a great juncture in economic history – the point at which human capital has overtaken both physical capital and natural resources as the primary source of prosperity and wealth creation. Previously it was ownership and superior management of either physical capital or natural resources that generated wealth. This, in turn, determined how human capital was managed and rewarded. Now that human capital has become the primary source of prosperity and wealth creation, the reverse is true. It’s the superior management of human capital that determines rewards in the marketplace.
In addition to this reversal of roles, there is one enormously important distinction between the current knowledge era and preceding eras. Unlike physical capital and natural resources, human capital cannot be ‘owned’ by employers. Consequently, an enormous premium now accrues to those organisations that have developed a superior capacity for managing human capital. In fact, a growing body of research points to human capital management as the single most important predictor of an organisation’s ability to outperform its competition.
The seachange that’s taking place as developed economies continue to become more knowledge-intensive is requiring senior executives to focus on human capital management as never before. This, in turn, requires measurement systems to help them optimise their return on people. The current state of human capital measurement within most organisations, however, typically consists of some (mostly unhelpful) combination of the following:
• Traditional HR metrics (which are, at best, lagging indicators of business results and, at worst, have nothing to do with business results) that have simply been re-labelled as ‘human capital metrics’.
• Employee satisfaction surveys with no known connection to business results.
• Simplistic balanced scorecard measures, such as percentage of managers that have been through a development program (while these measures are intended to be forward-looking and predictive, few actually meet these criteria).
• Historic ROI measures (too often designed primarily for budget justification, and typically unable to provide any guidance for how to improve an organisation’s return on people).
Thoughtful executives are desperate to change this state of affairs and to begin using a new generation of human capital measurement systems, which can actually serve as a guide to organisations seeking better business results through the improved management of their people.
Essential attributes of human capital measurement
There are six key attributes that a human capital measurement system must possess if it’s to be of maximum use to executives in managing and deploying people effectively within an organisation. The system must be:
1. Descriptive. At a minimum, a measurement system should produce summary statistics that provide a clear and succinct summary for each issue of interest. Descriptive data tend to focus on the occurrence of a phenomenon, its frequency or its intensity. For example, descriptive statistics can help an organisation monitor the degree to which an important best practice is (or isn’t) actually being implemented throughout the organisation.
2. Credible. A measurement system must be designed to provide the credible and unbiased insights needed to improve business results. Typically, any system designed primarily for the purpose of self-justification is quickly seen as suspect and is given little credence by senior executives. Many ROI initiatives, for example, fall into this category.
3. Predictive. A measurement system must produce statistics that help an organisation predict where it’s headed. Predictive measures are those that have been linked to the organisation’s desired business results.
4. Detailed. The information produced by a measurement system must be sufficiently detailed and disaggregated to provide insights into where action should be taken. For example, many types of information must be available across departments or business units in order to allow for a possible intervention to be targeted on those areas where it might be most successful.
5. Actionable. A measurement system should focus on those issues over which an organisation can exert influence. Other items, however interesting they may be, are unhelpful in enabling action to drive business results. The best example here is a counter-example: one well-known measurement system, the Gallup Q12, measures whether or not employees “have a best friend at work”. While this is indeed an interesting descriptive statistic and might even be predictive and detailed, it’s not an actionable piece of information and hence should not be an area of focus within a measurement system.
6. Cost-effective. As important as a powerful measurement system is to a well-managed business, it must be cost-effective if it’s to be sustainable.
Key human capital measures
Research has shown that there are five categories of human capital management factors that consistently predict business results. These represent a core set of measures that organisations should track through their human capital measurement systems. The five categories of measures are as follows:
1. Leadership/managerial practices include managers’and leaders’ communication, performance feedback, supervisory skills, demonstration of key organisational values efforts and ability to instill confidence.
2. Workforce optimisation includes an organisation’s success in optimising the performance of its workforce through the establishment of essential processes for getting work done, provision of good working conditions, strong hiring decisions and emphasis on accountability.
3. Learning capacity includes an organisation’s overall ability to learn, innovate and continually improve.
4. Knowledge accessibility includes an organisation’s ‘collaborativeness’ and its capacity for making knowledge and ideas widely available to employees.
5. Employee engagement includes an organisation’s capacity to engage, retain and optimise the value of its employees.
The best way to measure these factors within an organisation is through a thoughtfully-constructed survey of employees. Note that such a survey varies considerably from the typical employee satisfaction survey.
Using statistical techniques (such as those that are briefly described in the following section), each of the five human capital categories, as well as their specific components, can be linked to a variety of alternative measures of business outcomes. Such linkages provide senior executives with a clear prioritisation – a roadmap of sorts – of the human capital management initiatives that will generate the greatest improvement in business results.
Using human capital information
A few basic statistical techniques can be employed quite effectively to link human capital measures to business results in a way that is both credible and actionable. The central challenge in doing so is to isolate how one factor, such as training investments or the effectiveness of managers’ communications, causes another factor, such as sales productivity or safety, to change.
After gathering information on a given set of factors through a measurement system and employee survey of the type described above, the effects of each factor can typically be isolated by applying the principles of a quasi-experimental design. This identifies and quantifies causal relationships between inputs and outcomes.
Consider for a moment a drug dosage experiment where patients, all of whom have the same ailment, are given varying dosages of a medicine. The severity of the patients’ illness varies and they are each different from one another in a variety of areas, such as gender, weight and age. A statistical technique called regression analysis can be used to isolate the effect of the medicine on the outcome of interest (improved health) while controlling for both the variations in the dosages of the medicine (the input of interest) and the effects of the confounding variables of gender, weight and age.
Similarly, the effect of human capital management inputs on business outcomes can be isolated by controlling for the effects of confounding variables that affect different parts of your organisation in different ways (such as age of plant and equipment, local economic conditions and exchange rates).
This can be done by identifying and making use of the ‘natural experiment’ that exists within every organisation. Suppose your organisation has 25 sales offices (or factories, branches or locations), and that you have comparable ways to measure outcomes (such as sales per employee or safety) available for each office.
By regressing the input measures (such as those that might be measured through an employee satisfaction survey or by documenting varying ‘dosages’ of training) on the outcome measures, it’s possible to determine the magnitude of the effect (and its statistical significance) of each input of interest on the outcome of interest after eliminating the confounding effects of other factors that also affect outcomes.
Your ability to estimate such impacts with statistical precision will increase with the number of different units that are being analysed. In addition, the capacity to eliminate the effects of confounding variables is enhanced if outcomes (and inputs, if possible) are observed on more than one occasion for each unit. This allows for the ‘differencing out’of the effects of unit-specific confounding factors.
When the number of units – for example, sales offices – isn’t large enough to support regression analysis, alternative statistical techniques are available, such as differences of means (t-tests) and correlation analysis (Pearson coefficients). Or you might consider using a different unit of analysis. Instead of using outcomes for each sales office overall, you could use individual managers’ or employees’ outcomes, which could includ absentee rates, turnover or sales productivity. This would significantly increase the number of units available for analysis, thereby vastly increasing your ability to identify and isolate effects.
The statistical techniques described here can also be used to analyse the circumstances under which the impacts of interest are particularly large or small. For example, certain types of learning interventions might be effective only in units with high scores on learning capacity. This analysis generates insights into why the results are what they are, and as such, is useful in determining what actions should be taken to optimise organisational performance.
In sum, it’s possible to rigorously quantify how an organisation’s management of human capital effects business results when multiple (disaggregated) observations of both outcomes and inputs exist. The type of analysis outlined above essentially combines the strengths (and eliminates the weaknesses) of the balanced scorecard, employee surveys and ROI analysis.
If you have training investment data and/or can undertake a thoughtfully constructed employee satisfaction survey (as described in the previous section), techniques are readily available that will provide executives with the information they need to improve business results through enhanced human capital management.
Laurie Bassi is CEO of McBassi & Company and Daniel McMurrer is the vice president for research at McBassi & Company, a workforce strategy and benchmarking company. Email: [email protected].
