Twenty-One levels of “Government” in Product Quality

01 Nov 2013

To understand testing requirements you must first understand who is requiring it.

You may be familiar with the three branches of U.S. government: Legislative, Judiciary, and Executive. However, did you know that there are 21 branches that govern product safety and quality, and they can all impact the testing you may need to do?

If your reaction is: "That explains a lot!" then keep reading. To truly understand how a testing process becomes the rule of the land, we'll need to start examining a few more branches.

Non-government organizations (NGOs) cover a wide range of activities in this area, but for this discussion we will limit our focus to just a few, such as Standard Writing bodies, Auditing agencies, and trade organizations.

Standard writing bodies are organizations such as the Society of Automotive Engineers (SAE), the National Fire Protection Association (NFPA), and the American Society for Testing and Materials (ASTM) that agree to write test methods following certain established rules. Rules include allowing participation from all stakeholders, proper voting, and transparency. Standard writing bodies also agree to be audited by another type of NGO. Organizations including the American National Standards Institute (ANSI) audit the Standard writing bodies to ensure that they are drafting documents and voting in a fair way that complies to international policies like the International Standards Organization (ISO).

To recap, we have 12 levels of elected product safety government, plus NGOs who either write standards or audit the standard writers. There actually are two more hidden: the stakeholders (usually manufacturing companies) and individuals who volunteer (usually with the support of the company they work for). This brings us to 15 governing bodies in product quality and safety.

Once we have a test method written and published by a standards writing body, it does not yet mean it is a requirement. These organizations write a lot of standards and test methods that never become a requirement. To see how a standard test method becomes a requirement, we need to add a few more steps.

When we talk about what is "required" we really need to know "required by whom". Government regulations (Federal, State and Local) are an obvious source of requirements: These are laws. It's interesting to note that laws involving test standard are generally adopted from the standard writing bodies. For example, the Federal Motor Vehicle Safety Standards (FMVSS) rules for head lamps are based on the SAE test methods. The National Electric Code (which is adopted by State and Local government) is written by the NFPA.

There are other requirements I call "pseudo laws", that are not required by an elected body, but common practice within an industry or required by a trade organization. From a practical standpoint if 90 percent of a market or industry uses a particular method (such as the Air Conditioning, Heating and Refrigeration Institute's ranking for HVAC efficiency) then pretty much everyone needs to do this. Or if no major retailer will stock an item without a safety label then it might as well have the force of law. We just added Trade Organizations and Retailers to our branches, bringing us to 17.

Another set of requirements are called differentiators. These requirements are not written into law, or driven by industry practice(s), they come about because of the value proposition of the individual businesses marketing a product. Everyone tries to meet the laws and pseudo laws, but differentiators help a company compete in the marketplace. These requirements come about, not from outside a company, but from within and are driven by the value the company wants to have (or claim to have) for its customers. These requirements sometimes show up as trade organization tools or even government-sponsored options (such as Energy Star) which can provide a performance or quality differentiator. Who's driving this type of "requirement"? The manufacturer themselves. Now we're at 18 levels of product quality government.

But we still are not done. At this point we have 18 interacting branches that result in test requirements (laws, pseudo laws and differentiators). But what about the test itself? And is it just the test? Or do we need proof of the test and passing results?

Testing can be executed in a couple of ways. (1) an in-house lab follows the standard or test method and determines a result. In some case this is allowed by the Law or Pseudo law. Or (2) the testing can be done by a third party test laboratory. A third party lab will actually bring more levels of "government". There is the test laboratory that must follow ISO 17025 rules of an accreditation agency. ISO 17025 is not a test standard but is similar in that it's written by a committee who follows a set of rules, and ISO is audited to verify compliance to those rules. The accreditation agency will also operate under a set of rules (ISO 17065) in conducting the audit of the laboratory and it too will be audited. So there is the laboratory, the auditor, and the auditor's auditor. Three more levels bring the count to 21.

Now that you can see this interconnecting web of laws, pseudo laws, and differentiators, how they're written and adopted, and have an idea of who does what, can you begin to imagine how this impacts your quality and safety program on a day-to-day basis and the complexities involved?

Send your thoughts to me or reply below to this "21 branches of government" model. Stay tuned for my next blog which may help you decide exactly what type of quality and safety testing your company will need (hint: the answer is not necessarily what you'd expect).