Achieving quality step by step

Ever since we started outsourcing a lot of our work overseas, many companies have been real big on standard operating procedures. I think the theory is that breaking down your production process into a simple step-by-step operation makes it possible for even the most untrained worker to perform. While that can work well at a very basic level for those eager but inexperienced developing-world types, it hampers the ability of us still working on American soil to find creative ways to screw things up.

About ten years ago, the rage in corporate quality movements was something called ISO 9000. The idea was that if you documented (or “wrote down”) all your processes and then operated as you said you would, nothing could go wrong. No variation was possible when humans were turned into mindless, instruction-reading work-bots. Errors in this system were supposed to be so few that a special numeration system had to be devised to describe how tiny the odds of failure were. This was the concept of “Six Sigma”, or six mistakes out of all the fraternity or sorority members in the world.

Though ISO 9000 is still followed in some corporate backwaters of the world, it gradually lost credibility in the U.S. First there was the problem that even if American workers could make sense of the instructions, there was no guarantee that just because something was written down that it would work (see the 2008 Republican platform and any MapQuest directions for just two examples). And then there was the problem with the name of the initiative itself: ISO stands for International Society for Obduration, which I think has something to do with pity, and the 9000 part represented the year in which actual gains from the program could be seen.

The remnants of this system that still exist in most lines of work are now called “standard practices”. They used to be called “best practices”, but that was considered too elitist, I guess, and it was judged more important that we do everything the same, whether it was actually good or not. Now, whether the person doing the work is in Boston or London or Hong Kong or Neptune (in the year 9000), all they have to do is go to the corporate intranet, access the development and training section, then go to the operations page, then find the kind of process they’re doing, then call up the appropriate requirements, then find the “SP”, then start looking for another job because they missed a critical deadline while monkeying around on the computer.

When you do have time to follow the standard practice, you better pull up a chair because it’s typically going to take a while to get through it. One example I’m looking at breaks a particular operation down into 15 steps, which seems almost manageable until you consider that step 8 alone includes four checkboxes followed by 16 bullet points and six sub-bullet points. Other steps are ridiculously simple, like step 15 which involves taking your page off the printer. The standard practice doesn’t tell you how many fingers to use to pick up the sheet of paper, whether to use your left hand or your right hand or what kind of protective gear you should be wearing but, as the website warns all users, “don’t use a hard copy of these instructions because they are constantly being revised in the spirit of continuous improvement.”

When despite the best efforts of the quality mavens something wrong does make it out to a client, an investigation into how this could possibly happen usually takes place. A “service recovery account” is requested of the offending manufacturing site who attempts to figure out, usually several weeks after the error was committed, what step in the flawless process was not followed. Usually, the answer is something like “we didn’t work on this job”, and the matter is referred to another location. Once the site is definitively determined, the managers there will “drill down” through a massive collection of archived paperwork to figure out which individual or team was responsible (the drilling is just a figurative term at U.S. offices but involves an actual boring device for workers offshore). A corrective action is implemented, typically a scolding email to anyone who might’ve participated in the misdeed. We’re able to report back to the client that we appreciate they’ve pointed out an improvement opportunity that has made our process even better, and that someone won’t be getting their merit raise, if it’s ever decided these will be reinstituted.

What all this ignores is that some of the steps in a process are more critical than others, and that it takes an experienced person to know when it’s safe to cut corners and skip something trivial. If sub-step 2.4.7(A)(e) involves hopping on one foot while you key in your job number, you’ll see the Bombay skyline compliantly swaying with tremors while in Atlanta they’ll just take a chance they can skip the hopping. Our overseas workers are extremely good at doing exactly what they’re told to do, knowing they could be out on the streets if it’s found they cut a corner. At best, there will be “stand-ups” (where a top manager stands up before the group and yells at them), “letters of retribution” inserted into personnel files and, worst of all, week-long reprogramming regimens that involve the south Asian equivalent of a forced march. Virtually no one gets dismissed for cause domestically, since downsizing is certain to eventually take care of them anyway.

There’s a pendulum of emphasis that swings back and forth between quality and meeting deadlines that American workers seem to be better at timing. We’re much closer to the screaming customer to be able to tell when we’re about to enter a new era. We use those all-American traits of innovation and intuition and poor reading skills to perform from the gut what we think needs to be done rather than what some piece of paper says. And we can tell when it might be a good time take a lunch break to avoid those managers who are shocked (shocked!) to learn that a standard process wasn’t followed step by ridiculous, excruciating step.