Applying Design for Six Sigma Tools in the Hiring Process

Published: Tuesday, December 16, 2008 - 15:08

The cost of doing business is rising dramatically. Included in this is the cost to recruit and hire talent to join your team. It’s estimated that the total cost of one hire can exceed $30,000 when relocation and set-up expenses are included. That cost is minimal compared to the lost opportunity cost or the potential damage that can occur if the wrong candidate is hired.

Similar risks are present when promoting an employee into a position of greater responsibility. Frequently, the mistake is made of assuming that an employee who is successful in a particular position will automatically have similar success in a different position. This can be true; however, when different roles have different needs, a different skill set is often needed.

You can minimize the risk of making a poor decision and improve your probability of success. This article outlines simple but powerful tools that can be easily and quickly applied to many situations. A real-world example of how the Mercury MerCruiser division of Brunswick Corp. utilized design for Six Sigma (DFSS) tools to hire the right candidates for their engineering apprentice program is shown, and this example provides step by step instructions on the use of these tools.

Background
Design for Six Sigma is a disciplined methodology with a collection of tools to ensure that products and processes are developed systematically to provide reliable results that exceed customer requirements. A key function of DFSS is to understand and prioritize the needs, wants, and desires of customers and translate those requirements into products and processes that will consistently meet those needs. The DFSS toolset can be used in support of major new product development initiatives, or in stand-alone situations, such as the one described in this article to ensure that proper decisions are made.

Mercury MerCruiser utilizes a skills-based analysis method to interview and evaluate candidates for various positions. Although this process works well to understand generalities about candidates, it has the potential to leave gaps in understanding how various candidates compare to one another with respect to key factors that affect fit and performance. DFSS tools were used to augment the generic human resources analysis method and to customize the process for a particular situation.

Mercury MerCruiser participates in a program in conjunction with the U.S. Department of Labor by which employees go through an intensive four-year engineering apprenticeship program. The program is designed to include classroom and on-the-job training requirements, and the end result is a well-rounded engineering technician who has obtained a nationally-recognized certification.

To improve its ability of identifying the characteristics, skills, and experiences desired in potential candidates entering into the engineering technician apprenticeship program, MerCruiser utilized specific DFSS tools that facilitated the hiring process.

Application of DFSS tools
First, the recruiters had to determine what the expected outputs of apprentices are. They also wanted to identify what characteristics and experiences would be valuable inputs to the apprenticeship program. Because the main customer base was small enough, the voice of the customer was gathered in an informal focus group setting and recorded in the form of a supplier, input, process, output, customer (SIPOC) diagram. A SIPOC diagram is a simple but powerful tool that can be created in a very short time-frame to outline the boundaries of a process, capture what the expected outputs of the process are, and who receives those outputs. It also identifies the inputs and suppliers required to execute the process. Other tools that can be used to gather and understand voice of the customer include surveys, interviews, Kano analysis, KJ analysis and quality function deployment, or house of quality. The combination of tools you choose depends upon the complexity of your initiative.

Once recruiters determined what to look for in a candidate, they wanted to understand the relative importance of each input. To do this, they utilized a pair-wise comparison based on near consensus rankings. A pair-wise comparison forces the participants to rank each item against the others in a series of A-B comparisons. While several items may be deemed critical, this tool forces the team to choose which items have a higher priority (even if the relative difference is negligible).

This tool is excellent for small groups (five people or less) when all team members willingly participate. One risk with this tool is that a dominating personality may unduly influence the outcome. If that starts to occur or if you see a pattern of a team member or two not speaking up, you may need to use some method of hidden voting so all opinions can be understood.

For larger groups, other tools are recommended to ensure that everyone’s opinion is heard. One method is for each participant to do a pair-wise comparison individually, and then combine the rankings from all of the participants. Another tool that can be used to prioritize needs is a multivote technique. With this tool, participants get a certain amount of votes to use on the various options. They then assign points to each option based on their perception of relative importance. Oftentimes, it’s a good idea to set a maximum number of votes that an individual can use for one item to ensure that a strong opinion doesn’t skew the results. This is a good method to quickly understand the factors that are critical and weed out the less important minutia.

Here is how the pair-wise comparison for this example turned out in the case of Mercury MerCruiser:

By utilizing a Pareto diagram, the recruiters were able to group the characteristics by relative importance. They used a 1, 3, 5, 7, 9 rating system with the higher number being of greater importance to allow for delineation between the characteristics.

Once the characteristics were prioritized, they utilized a Pugh matrix to determine the relative rating of each candidate with respect to each characteristic. Again, they used an odd number 1–9 rating system (with negatives and positives) to rank how each candidate compared to the others on each characteristic. They then chose one candidate that they were familiar with as the baseline, and the +/- 1-9 ranking system allowed them to compare against this baseline candidate and against all other candidates at the same time.

A Pugh matrix is an excellent decision-making tool when there are various concepts or options that need to be evaluated against one another, based on multiple requirements or items. Benefits include identifying the relative strengths and weaknesses that each option has. Where applicable, this can be an excellent tool to develop hybrid solutions to complex problems as the team draws from the strengths of various concepts.

When using a Pugh matrix, you should be aware of biases that people bring with them when completing the matrix. If someone on the team knows which factors have the biggest effect on the decision, they may artificially or intentionally inflate the scores on those issues for the concept they personally favor. To minimize this, it’s best to randomize the decision criteria and hide the importance factors and the total scores. Here is the Pugh matrix shown with the scores hidden.

Hiding the relative importance and the total scores allowed the recruiters to have an open and honest discussion about each candidate, and they could share their perceptions of how the candidates relate to one another with respect to each characteristic. Once the scores were agreed upon, one of the recruiters revealed the relative importance of each factor, as well as the total score for each candidate.

It should be noted that a Pugh matrix may not be a definitive tool (i.e. you may not necessarily choose one concept just because it has a slightly higher rating than another). The intent is to quickly differentiate the stronger concepts from the weaker ones, and to give a directional view of the options. In the case of this example, the top two candidates clearly stood out among the group.

As the recruiters worked through the process, they continually went back and reviewed the tools that were utilized earlier. There should be a direct link between the SIPOC (or whichever voice of customer tools are used), the prioritization tools, and the decision-making tools. In this case, all of the inputs of our SIPOC were evaluated in the pair-wise comparison and then used in the Pugh matrix to identify the top candidates.

As with most of the DFSS tools, the value isn’t in the tool itself, but in the interaction of the team and the flow of knowledge that comes from working through the exercise. In this case, the generic human resources tools used in the hiring process didn’t touch on the specific needs on the particular position. This would have limited the team’s ability to fully understand what characteristics gave the candidates the highest potential for success, and it would have also limited the evaluation of the candidates. The tools in the DFSS methodology are an excellent way to facilitate the exchange of information, and by expanding their use to other business processes such as the hiring process, you can improve many facets of your business.

About The Author

Barry Johnson

Barry Johnson serves as president of Performance Optimization Associates LLC, a performance improvement consulting group. Johnson has a bachelor’s degree in mechanical engineering from the University of Tulsa and a master’s degree in leadership from Grand Canyon University. He has more than 25 years of operations, engineering, and management experience in diverse industries such as oil and gas, automotive, consumer goods, electric utilities, and recreation. He can be contacted at barrygjohnson@sbcglobal.net.