Product Failure – What to do when it happens to your company

A long time ago, after a disappointing personal failure, someone told me: “Steve, there are no failures, there are only results. Sometimes you get what you want, and sometimes you get what you need—specifically, information about how to get what you want. Both results are helpful.” This changed my whole attitude on life. Instead of becoming emotionally distraught about failure, I turned it around and start seeing it as something valuable—information about what I need to do the next time differently.
Every company wants to sell a product that never fails. The reality is that even great companies have product failures. The difference between the great companies and the good ones is not perfection, but how they view failure and what processes they use to address the failure. Do they see it as an opportunity to learn or as a reason to point fingers at others? The choice that each company makes will depend on how Management reacts to failure.
Having a formalized procedure for product failures can go a long way in helping the team embrace the idea that the fault is an opportunity to learn and act, which will create positive results in the future. Having a formalized product failure procedure normalizes the inevitable failure and provides an effective mechanism for change.
What should this process be? Of course, each company is different, but keep in mind the following four key areas:

What went wrong:

It’s essential to understand all the circumstances of the failure. Collect as much information as is practical. Details are crucial to be able to reproduce the crash in the lab. Avoid jumping to early conclusions before all the data has been collected.

Why did it go wrong:

In engineering speak, this is called “Root Cause Analysis.” It is essential to understand that you need to discover the most fundamental issue—the first mover as it is often called. For example, saying that Q1 failed is not enough—we must know why Q1 failed. A better answer would be Q1 failed due to overcurrent because of tolerance stack up issues that resulted from an inadequate design checklist.

What will we do to make sure it does not happen again:

The QA guys call this “Corrective Action.” What actions are you going to take to make sure this type of failure does not occur in the future? For example, a Corrective Action might be to add a procedure to the design process that includes Moto Carlo analysis to understand the tolerance stack up and its effect on component stresses.

Follow up with the customer:

No customer has a right to expect perfection—we are only human. However, they can expect that you continuously improve. In today’s world of public knowledge, reviews, and the Internet, you cannot afford to have any bad press. Follow up with the customer and explain what you have done to make sure the problem they experienced will not happen again.

Comments are closed.