Product Testing: What You Ignore Could Kill Your Business
We've all done it at one time or another in our careers. You are well along in the development cycle of a new product. The sales group gets a great opportunity for the new product that includes a requirement for an environmental specification that was not explicitly a consideration of the design. It's too late to change the design. Your best engineering instinct says the product will work. However, there is no money left in the budget for testing, and there is no specific certification agency or contractual oversight to require testing. What do you do? In many of these cases, the design is completed, the data sheets are written, and the product is sold with the hope that there will not be any latent problems in the field later.
Unfortunately, I believe that this and many other similar stories are all too common in the display industry. The example may be an integrator that cannot afford the cost of re-testing, so they assume the stated compliance on the display sub-assembly they are integrating covers the higher level system. It may be a specific component such as an inverter or graphics card that has to meet a series of generally accepted market specifications, but the cost to test each combination of each extreme is just too high, so the supplier tests a subset of the features and relies on the market to find any subtle issues for them. Whatever the circumstances, the rigor of thorough product testing is frequently sacrificed in favor of time to market and cost management.
I have not seen any useful statistics on this problem, but I also know from my industry experience that occasionally it becomes a serious problem for both the customer and the supplier. Not knowing the full extent of the capabilities or limitations of your product creates a significant business risk. Rarely is there any malice intended, but the back-end costs to recover from unknown performance deficiencies can be devastating for a business relationship. The problem is sometimes exacerbated by the commonly used practice of "self-certifying." Although limits on electromagnetic emissions and susceptibility performance are legally mandated in almost every market in the world, more and more government entities are allowing product manufacturers to self-certify, simply by labeling the product as compliant and maintaining a technical file of the test results available for public inspection if ever needed. In both the United States and most of the European Union, there is no official channel that requires submission of the test results and no license to be issued upon approval. It is therefore very tempting to simply design around best practices, label the product as compliant, and hope to not have a problem in the field. The cost savings from not performing EMI testing alone can be significant, from $10,000 to as much as $50,000 per product in some cases.
However, the cost savings may be a Pyrrhic victory. Even if the test avoidance does not include any legally mandated items, it does mortgage the future for your company's reputation and could create serious financial liabilities down the road. Premature field failures, unexpected interference with other equipment, safety problems, etc., can all be very expensive to recover from and, if they are severe enough, may spell the end of your chances for success in a given marketplace. Therefore, it is not a wise business decision to forego prudent and comprehensive product testing, regardless of the motivations. Moreover, avoiding the tests related to regulated items such as EMI or safety, regardless of self-certification rules, is in most cases illegal.
Stephen P. Atwood