Designing Best Practice

Our best practices research explored how talent mobility experiences differ across high-performing organizations. We categorized high performance in two ways – “Business Performance” and “HR Best in Class.”

Business Performance: We identified organizations that had performed in the top 25% of our sample audience who, in the last 12 months, had demonstrated higher revenue growth, total revenue, revenue per employee, percentage of market share, net promoter score, and sales growth year-over-year. We asked HR Leaders to rate how several business metrics have been impacted over the last year (i.e., decreased, stayed the same, increased across revenue growth, revenue per employee, percentage of market share and sales growth year-on-year).

The organizations that were classified as top 25% in Business Performance had the following characteristics:

• The majority come from Professional, Scientific, and Technical Service industries (26%);
• The majority are large organizations with 10,000+ employees (35%);
• The sample includes organizations in North America, Europe, Asia, and the UK

HR Best in Class: Next we segmented our respondents according to how several HR metrics have been impacted over the last year (i.e., decreased, stayed the same, increased). Based on their responses, we were able to identify organizations that performed in the top 17% of our sample in terms of HR Metrics. In the last 12 months, these organizations reported less voluntary attrition, higher internal mobility, lower recruitment costs, fewer unfilled positions, higher employee productivity, and increased spending on training and development.

The best-in-class HR practice group had these characteristics:

• The majority come from the Professional, Scientific, and Technical Service industry (28%);
• The majority are small organizations with less than 999 employees (72%);
• The majority of spending is between USD 500-1000 a year per employee on training and development.
• The sample includes organizations in North America, Europe, Asia, and the UK

Our research found a strong link between “HR Best in Class” and “Business Performance” (r =.443** p < .001)

Organizations in our sample that ranked as “HR Best in Class” (i.e., they had lower voluntary attrition, higher internal mobility, lower recruitment costs, fewer unfilled positions, higher employee productivity, increased training, and development spending) also outperformed organizations in business performance (i.e., they had higher revenue growth, total revenue, revenue per employee, % of market share, net promoter score, sales growth year-over-year).

This finding provides a compelling reason for organizations to pivot their focus to implementing the tactics and focus areas highlighted in this report to drive enhanced HR practices – leading to increased business performance. Our key findings about the “HR Best in Class” group of organizations included some insights on what was being prioritized in their business.

The Critical Levers to “HR Best in Class”

Our research found that the “HR Best in Class” group were investing in some critical levers to drive better performance on HR metrics across three audiences – the organizational lens, the leader lens, and the HR lens. Investing in employee development was strongly correlated with best-in-class organizational performance (r=.264**), as well as enabling leaders to support employee development (r=.301**), while HR were enabled to have talent and skills intelligence to have visibility to bench strength (r=.282**).

Survey Design

The Fuel50 Global Talent Mobility Benchmarking Study explored current talent mobility practices, best-in-class talent mobility, and the imperatives for talent mobility in the future from two different angles:

1. One section of the survey was designed specifically for HR Leaders and contained a mix of 147 multiple choice and Likert style questions.
2. The other section was for Employees and included 35 questions.

We invited both HR Leaders and Employees to complete the survey via social media posts, direct email reach outs to Fuel50’s valued community of clients and thought leaders, and by asking attendees at our virtual FuelX Conference in April 2021 to respond.

Insights on Respondents

Within the sample of HR Leader respondents, the largest proportion worked for organizations within the industries of Professional, Scientific, and Technical Services (25%), followed by Finance & Insurance (9%). 37% of the HR Leaders were from organizations with over 10,000 employees. The majority of respondents organizations had a presence in North America (74%), followed by Europe (44%) and Asia (42%).

Within the sample of Employee respondents, 41% have been in paid, full-time employment for over 20 years, with 35% identifying as being in an Individual Contributor/Specialist role and 29% as a Senior Specialist. Most respondents (41%) worked in the industries of Professionals, Scientific, and Technical Services.

In terms of tenure, 31% have worked in their current organization for 1 – 2 years. 22% have worked at their current organization for 3 – 5 years. More than half (53%) of the sample were in North America, with the remainder located in Australasia (Australia and New Zealand) 17%, Europe 11%, Asia 11%, and the rest of the world 9%. The majority of respondents worked in an organization with less than 100 employees (39%), with 22% working in organizations with between 110 – 999 employees.

Data Cleaning Process

To ensure the quality of the data and statistical validity of the study, we used several data screening techniques.

• First, we examined whether our sample participants fit the requirement of the population we intended to capture, which was HR leaders and employees who are currently working in an organization. There was one employee who identified as unemployed with a previous tenure less than one year, and thus they were removed from the employee sample.
• Second, missing data was evaluated at the person level, construct level, and item level. According to Downey and King’s (1998) recommendation, missing data at a rate of more than 30% potentially raises concerns. There were no respondents with more than 30% missing data, so no responses were removed at this data cleaning stage.
• Third, we inspected response times that were exceptionally fast. These can indicate a lack of effort or attention, with a few studies stating that exceptionally fast responses are assumed to be careless. Huang et al. (2012) suggested a logical cut-off of 2 seconds per item criteria. No response times were less than the cut-off criteria, so no responses were removed from the sample using this data cleaning technique.
• Lastly, a statistical screening technique, Mahalanobis Distance (D) statistics (Mahalanobis, 1936) were generated to compare each respondent’s score to the sample mean across all responses for all the items in the survey. A response vector identical to the sample mean score of the item has a Mahalanobis D of zero, while high values of D indicate an extreme deviation from the sample means across the survey responses, thus, flagging the response as potential low-quality data (Meade & Craig, 2012). This statistical analysis identifies any extreme deviation from a normative response pattern which could indicate multivariate outliers or low effort responding (DeSimone et al., 2015). It is important to note that extreme outliers may influence the mean and increase the variance of survey items, and thus researchers are recommended to screen out responses at the top or bottom 0.1% of the chi-square distribution (DeSimone et al., 2015; Field, 2018). We identified 2 cases from the HR leader sample and 1 case from the employee sample that exceeded the thresholds for extreme Mahalanobis D. We removed these suspicious cases as they could be the source of low-quality data, noting that extreme outliers can pose a significant risk to the reliability and validity of the subsequent statistical inferences produced by a study. By the end of data cleaning stage, 2 cases were removed from the HR leader sample and 2 cases were removed from the employee sample.

References

DeSimone, J. A., Harms, P. D., & DeSimone, A. J. (2015). Best practice recommendations for data screening. Journal of Organizational Behavior, 36(2), 171-181. https://doi.org/10.1002/job.1962

Downey, R. G., & King, C. V. (1998). Missing data in Likert ratings: A comparison of replacement methods. The Journal of General Psychology, 125(2), 175-191. doi:10.1080/00221309809595542

Field, A. (2018). Discovering Statistics Using IBM SPSS Statistics (5th ed.). SAGE Publications.

Huang, J. L., Curran, P. G., Keeney, J., Poposki, E. M., & DeShon, R. P. (2012). Detecting and deterring insufficient effort responding to surveys. Journal of Business and Psychology, 27(1), 99-114. https://doi.org/10.1007/s10869-011- 9231-8

Mahalanobis, P. C. (1936). On the generalized distance in statistics. Proceedings of the National Institute of Sciences of India, 2, 49-55.

Meade, A. W., & Craig, S. B. (2012). Identifying careless responses in survey data. Psychological Methods, 17(3), 437-455. https://doi.org/10.1037/a0028085