Finance executives may sense that the more they’re involved in hiring and employee-development decisions — as opposed to giving human resources a lot of say — the better those decisions turn out to be. But new research points to a possibility that executives may not be fully aware of just how important their personal involvement is to the success of their organization.

Many companies track how hiring and development decisions work out (especially within critical talent segments), using measures such as performance-review scores and turnover rates. Few, though, expend much effort differentiating between the quality of talent decisions made by HR and non-HR leaders. That is probably a mistake, judging by a study from the Institute for Corporate Productivity (i4cp), a member-based research organization.

The participants, 390 senior corporate executives with various titles, were asked how effective their organization was in its use of 11 different strategic HR practices, and also the extent to which they performed eight HR-related measurements.

One result was a significant outlier. Only 15% said they specifically measured, to a high or very high extent (two of the response options on a five-step scale), the effectiveness of talent decisions made outside of HR. Among all 19 practices and measurements, that was the lowest rate at which one of the top two rankings was selected.

Why is that significant? Because the research also sought to determine whether, and to what degree, the effectiveness of the 11 strategic practices and the usage level of the eight measurements correlated to company performance. And the strongest correlation was found with regard to — you guessed it — measuring the quality of talent decisions made by non-HR leaders.

Think about that. The degree to which a company measures the quality of talent decisions by non-HR leaders is a better predictor of company performance than, for example, how successfully the company identifies where talent has the greatest potential for strategic impact. And it’s far more highly correlated with company results than how greatly the organization measures the impact of high versus low job performance generally, or of training or motivation programs.

Companies “should be jumping all over this,” says Jay Jamrog, senior vice president of research for i4cp. “It sure looks like it can be a competitive advantage, because few companies are doing much of it and it has the highest correlation with company performance.”

Now it strains one’s credibility to imagine that any CEO looking to hire a CFO, or any CFO bringing in a new controller or another leader of a major finance discipline, would cede to HR much of the responsibility for making the hiring decision. But depending on the company, there may be less vigilance when it comes to keeping watch over the development of potential future leaders a few levels down.

“A lot of executive teams don’t spend nearly enough time on people development,” says Irv Rothman, CEO of Hewlett-Packard Financial Services, a wholly owned HP subsidiary, and author of the new book Out-Executing the Competition (John Wiley & Sons, 2012).

“You have to pick out the best people and figure out how to move them along with the assignments and training that will make them capable of stepping up to [the] next level,” Rothman tells CFO. “That is just not something you can leave up to HR. It is on the agenda every month when our executive team gets together.”

Wonk World
To appreciate the foregoing information, you may not need a deeper understanding of the correlation between company performance and measuring the talent decisions of non-HR leaders. But for those who want that, here it is.

In addition to rating the success or prevalence of the 19 strategic practices and HR-related measurements, participants rated the performance of their company in four categories: revenue growth, market share, profitability, and customer satisfaction. The ratings were combined into a “market performance index.” (For statistics wonks: The index served as a “multivariant dependent variable.” For purposes of the following explanation, let’s just call it a “dependent variable.”)

A statistical-analysis program then determined that the extent to which a company measures the quality of talent decisions by non-HR leaders (an “independent variable” in this research) has a 0.18 correlation with the company’s market performance index.

Describing to those who aren’t statistics experts what a 0.18 correlation means, in this context, is no easy feat. (Wonks: To be precise, it’s a 0.18 correlation at the 99% confidence level.) It may be shown best by breaking down the explanation into its components, as follows:

• Imagine a graph representing the distribution of participant responses on how extensively they measure the talent decisions of HR leaders (i.e., the independent variable). The graph shows how many participants chose each response option within the five-step scale.
• Now imagine a similar graph depicting the distribution of the market performance index scores of the participants’ companies (i.e., the dependent variable). It shows how many companies are within different score ranges.
• A 0.18 correlation means that 18% of any tendency for the two graphs to match up in the direction of their up and down movements is based on a predictable statistical relationship between the dependent and independent variables. The other 82% is subject to other variables or coincidence.

(Note that a correlation is not the same as cause-and-effect. The latter requires a more complex statistical probing, such as a regression analysis.)

The more components a multivariant dependent variable (like the market performance index) has, the lower the correlation with an independent variable tends to be. Given that there are four such components in this case (revenue growth, market share, profitability, and customer satisfaction), a correlation of 0.18 is high indeed, Jamrog says.

Among the other 18 strategic practices and HR-related measures, 10 registered correlations of 0.11 through 0.14, and 1 each was at 0.15 and 0.16. That might seem like a narrow range, but in such research differences as small as two or three percentage points are quite significant. So, the correlation between market performance and measuring the talent decisions of non-HR leaders is unquestionably an outlier, according to Jamrog.