Where’s Your Weakest Link?
Alan Shepard, the first American astronaut, is quoted as saying “I was up there looking around, and suddenly I realized I was sitting on top of a rocket built by the lowest bidder.”
He didn’t say that because it made him feel more secure!
This came to mind recently as I was dealing with a dead computer. It’s a two-year-old name-brand machine (I’m not going to name names because it could have happened with any vendor) that my olive oil business DaVero uses in the shipping area. It behaved oddly for a couple of days, then just flat-out gave up the ghost.
I called the vendor’s “Small Business Support” line, because this was one of their ‘business’ computers, and because I’d paid extra to get next-day, onsite repair. To their credit, the hold time was relatively short. But it took well over two hours of my time to drag the well-meaning but frankly Neanderthal person on the other end of the phone through their scripts. Along the way, I had to run various diagnostics and then physically disassemble the computer, unplugging various components to try to isolate the issue. (It took almost 10 minutes to convince him that re-installing Windows wasn’t going to help, because the computer wouldn’t even boot.) At the end of the process, he concluded that the power supply was defective (which I told him up front), and that they would need to replace both it and the mother board.
“So what time tomorrow will your people be here?” I asked.
“It may take several days,” he responded.
“Um, what part of ‘next-day, on-site repair’ corresponds to ‘several days’?”
“Well, first we have to get the parts to the technician, which we may not have in stock, and then they have to schedule the visit.” (It has now been over a week, and we still don’t have a firm appointment for repair.)
I knew better than to argue. But I also know better than to buy from this vendor ever again.
Still, there are two key observations from this experience.
First, today’s computer vendors don’t ‘manufacture’ anything. At every price/performance level, they assemble their computers from components they buy from the lowest bidders, who in turn buy their parts from the lowest bidders. The entire production chain works on the basis of an expected (and, apparently, acceptable) failure rate(*). If, later, the components have an unacceptable failure rate, they buy from the second-lowest-cost bidder.
Second, it’s clear that most technology vendors are doing their best to push tech support out to the user as a way of controlling costs. If I hadn’t been there and my (very talented but technically challenged) wife had been forced to deal with this, she would have given up within minutes and just bought another (low-cost) computer.
The bottom line, of course, is that any failure is unacceptable for my business, so I need to be prepared for it. That means not just backing up my data, but also having a plan in place so that when Murphy points at us we don’t grind to a halt.
We got lucky three ways: I have a lot of technical savvy and was able to identify the problem quickly; we had another computer we could use; and all of our important data is in NetBooks, a system designed with “bank-grade” redundancy and security. But it’s entirely possible that we would have been unable to ship for the past week, or would have had to haul everything off to the UPS store – both alternatives that would cost us money and degrade our customers’ experience with us.
I suggest you look around your business and ask yourself what would happen – and what you would do – if a piece of equipment (or, for that matter, a person) were to stop working. I did, and it’s a sobering process.
You can’t afford to take all the risk out, of course, but my guess is you’ll find you’re taking risks you didn’t know about – and maybe don’t have to take.
(*)The name for this, by the way, is “MTBF”, or “Mean Time Between Failures.” It’s a way of quantifying the probability of failure. Saying that a hard disk drive has an MTBF of 40,000 hours doesn’t mean your drive is likely to fail after 4.57 years; it means that the average across a large group of identical drives will be 4.57 years. Yours might fail the first day, or run just fine for 30 years. Think of it as similar to the probability of getting struck by lightning: it may be low, but the statistic is irrelevant if you’re the one who gets hit!
