How many belts?

Q: My organization is building its efforts to implement lean Six Sigma. I would like to know if there is a rule of thumb about the ratio of certified experts or practitioners to employees in an organization.

Michael Washington
Houston, TX

A: When organizations began to wake up to Six Sigma in the late 1990s and early 2000s, some Six Sigma teachers offered rule-of-thumb ratios you are now questioning. Some suggested one Black Belt (BB) for every 100 employees, one BB for every 10 Green Belts (GB), or one Master BB (MBB) for every 10 BBs. And there are other rules of thumb that vary significantly.

I cannot say with 100% accuracy how these rules started, but I would imagine that one well-known organization—through empirical observation—determined this set of ratios worked well for it. Consequently, others holding this organization in high esteem carried forth these ratios unquestioningly and even taught them as gospel.

The use of ratios concerns me greatly. It tends to equalize individuals at each belt level. I wouldn’t assume any BB or MBB is capable of doing any project. Also, ratios implicitly assume each belt level does the same type of work. This is rarely true.

If you expect work content to vary—even if it involves the balance between project work and mentoring or coaching work—it might be better to address manpower needs while considering other criteria, such as geography and demographics.

Also, if ratios are established at the functional group, plant, site or department levels, they will be destroyed as soon as organizational changes take place—and rest assured, they will.

If you still feel you must make use of ratios, I suggest you consider a few things, the first of which is an adequate supply of meaningful projects. If you are developing belts from within, you will need a project for each belt going through training. An insufficient number of projects can severely hamper your ability to train.

Otherwise, if you have a sufficient number of projects, you can move on to the next steps to begin populating the belt levels:

Start with a job description. The description must be sufficiently detailed so it can be translated into hours against expectations. For example, the MBB is expected to work projects 60% of the time and to coach or mentor 40% of the time.

If there is an expectation that a BB is coached or mentored one hour per week, then one MBB can coach or mentor no more than 16 BBs per week. Therefore, an upper limit has been set. This approach will help you establish your manpower plan.

But you should ask yourself a few questions: Is this a reasonable number? How does 16 BBs per week fit in with other goals your organization may have set? Is it too slow? This same approach can be used for BBs, but you must decide whether you want to use it.

One problem with this strategy is the strong temptation to use a one-size-fits-all job description. But you might, for example, want to have one or two MBBs conducting training and another developing training. These individuals would not be covered by the previous job description and would be at a severe disadvantage when performance appraisal time comes. Instead, it’s better to develop job descriptions appropriate to the job.

Define the goals of Six Sigma. After your manpower plan is established, you can reconcile it against assumptions made against projected dollar returns for projects completed at the MBB, BB and GB levels. If your projected returns are insufficient to meet expectations, your manpower profile by belt level may need to be adjusted or perhaps shifted.

Again, this is a decision you must make. In fact, you might need to go through several iterations until you are confident you can achieve the goals of the organization. The bottom line is that the number of MBBs, BBs and GBs—like any manpower plan—must be planned based on a specified timeline that will help achieve the organization’s goals.

T.M. Kubiak
Performance Improvement Systems
Weddington, NC

Looking for meaning

Q: After several years working in the field, I can’t reach a conclusion about the difference between calibration and verification. I just want to understand—in plain English—the difference between these two terms because I often find several interpretations of the definitions given by the International Vocabulary of Metrology (VIM).

Rafael J. Mateo
Santo Domingo, Dominican Republic

A: Indeed, the VIM gives very general definitions of calibration and verification applicable to various situations.1

In plain English, calibration provides for trueness of the measurement results, while verification ensures the calibration is correct. Calibration and verification are subject to industry standards and regulations. The following explains the idea of calibration and verification in an example of clinical laboratory:

A simple test on a patient sample can include a reaction producing a chromophore with certain light absorption spectrum. The reaction signal ("indication" in VIM) proportional to the analyte concentration is a combination of the intensities of light transmitted through the reaction cuvette at various wavelengths.

The test result is obtained by multiplying the reaction signal with the calibration coefficient, which is calculated during instrument calibration as a ratio of the reaction signal and the analyte concentration in the calibrator.

The concentration of the analyte—for example, serum calcium—in the calibrator is established through a process of transferring value from respective World Health Organization Certified Reference Material. Usually, the instrument calibration is performed periodically by laboratory personnel to ensure test results are reproducible between instruments, reagents and laboratories.

Calibration verification ensures the calibrator value assignment is correct by testing samples of calibration verification material, with test results being within specified limits. There is ongoing verification of system calibration (instrument and reagent) via testing control materials in each analytical run.

If the test results of control materials are within specified limits, the calibration is verified, and the patient results are valid. If not, the system is recalibrated.

Because of the criticality of consistency for patient test results across laboratories and over time, there’s an ongoing process of periodical proficiency testing (PT) of the clinical laboratories with blind samples sent by the PT administrators. The results returned to the PT administrator for an analyte must be within specified limits for the laboratory certification to be renewed for the next period.

In other cases of measurements—for example, electrical quantities such as voltage—the concept of calibration and verification using reference sources of the quantity being measured is the same. The process is often much simpler, however, and the calibration is done and certified by authorized laboratories with required frequency.

Jeffrey E. Vaks
Roche Molecular Diagnostics
Pleasanton, CA


  1. ISO/IEC Guide 99:2008—International vocabulary of metrology—Basic and general concepts and associated terms is available at www.bipm.org/en/publications/guides/vim.html.

For more information

  • Grachanen, Christopher L., "True Meaning," Quality Progress, November 2011.

Average Rating


Out of 0 Ratings
Rate this article

Add Comments

View comments
Comments FAQ

Featured advertisers