Many products have a longer viable life than you might expect; are you and the user prepared for that?
Products with a “long tail” are often cited as good, steady moneymakers. After all, once your product-development and tooling costs are amortized — typically within a year or so — the margins on those products really go up. The idea is that you release the product, it catches on and production ramps up, it gets designed in or sells well (depending on if it is an OEM or retail product), and the revenue comes in nicely.
Then, after a few years, sales taper off, but some orders still come in. As all costs have been covered and your manufacturing costs are lower due to experience and other factors, there is a steady but dependable stream of revenue at high margins. We’ve even seen ICs that are still offered by their original vendors 10, 20, and more years later (although clearly not recommended for new designs) as well as ICs from sources such as Rochester Electronics, which specializes in making “obsolete” parts fabricated using tooling purchased from the original source.
That’s the good news. The downside is that the long life of major products can lead to a dilemma because the product is still in use or finds a new owner but the support isn’t there. This is a lesson being learned the hard way by the new owners of The Boston Globe. Shortly after the purchase of the newspaper (yes, newspaper), the Globe moved out of its historic headquarters that included reporters, editors, sales, marketing, art, and page layout staff, along with the behemoth, 60+-year-old electromechanical printing presses (see figure). The print operation was moved to a new facility some distance away, and the old presses were replaced with “new” ones that were just 20 to 25 years old. (There’s a fascinating video tour of the old presses here and some impressive statistics and photos here.)
The numbers associated with these ancient newspaper presses are hard to grasp: four stories tall, weighing 542 tons, handling 400 tons of paper per week, on rolls of paper weighing 1,400 pounds each. Source: Boston Globe.
The need for the newer presses was simple: While the printed paper is a declining part of the business, it’s still a needed revenue source, but those presses don’t need to run 24/7 to fulfill the print-run demand. So the Globe contracted to print other area newspapers, including their top rival, as well as do standard commercial printing of advertisements, flyers, and brochures. It all makes sense in principle, and the firm responsible for the new facility was quite proud of it, as seen here
But there’s a problem: The used presses that they bought were not easily re-configurable for different commercial print jobs (see here
). Unlike the non-electronic “dinosaur” presses that they replaced, the newer ones are software-controlled and user-programmable, but finding someone to work on that ancient code and the tools with which to do it has been a problem. The original vendor in Germany is looking for an “old-timer” who still remembers something and also wants a small fortune for the work (if they can even do it) because there is no long-term future in that type of upgrade investment. Long story short: The Globe is losing money because it can’t print its own daily newspapers reliably and on time, and they can’t take on outside projects with confidence, either.
The situation is somewhat analogous of the Y2K crisis that we faced at the end of the 20th
century. Some pundits predicted that “everything” would come to halt, while others said that the effects would be modest and manageable. The reality was in between, especially because there were apparently enough old-school FORTRAN and COBOL programmers still around to deal with the issues. Also, the Y2K problems had many disparate applications and thus a wider knowledge base, unlike the Globe’s printing presses, which are fairly unique.
The lesson here is clear: In many applications, products often have a much longer life and use than we think they will. Unlike the smartphone-centric consumer world with its two- or three-year product lifecycle, in the industrial, medical, instrumentation, and military world, a good product not only has a long in-use life, but spare parts and software maintenance are genuine concerns that must be considered.
At the very least, some solid, commented documentation is critical — and please, put it on paper, because the software-based or online document format that you use may not be viable in 20+ years
(maybe not even the Adobe PDF format). At least that way, you can say that you have done the responsible and professional engineering thing — that is, assuming that management will support you!
What’s the oldest product that you have ever had to try to maintain or even upgrade? And what’s the longest-living product that you have ever helped design?