After having spent far too much time out in the real world lately, I have finally had a chance to drop back into the blog and I am pleased see how ably Kevin has been fighting the good lean fight without me. I can't help but chime in on the outsourcing question, however. The folks who wrote about lean outsourcing framed the issue in a theoretically fine little chart, with cost, quality and lead times all put in their corners of an isosceles triangle and, as all good outsourcing discussions go, suggested that manufacturers seek to evaluate their outsourcing opportunities holistically - on the basis of all three implications.
I have a better idea: How about if manufacturers figure out what their cost really is first. Forget about the cost of lead time and quality. Most companies have no idea what it costs to take widget materials A,B,C and D; put them together into sub-widgets E and F; slap those together into a final widget G; then put that into a cardboard box. Never mind the higher level questions of how long it takes or how many they get right.
When companies cannot figure out the cost of running and taking care of the widget making machines, or the cost of moving widget parts around their factories, the esoteric costs of such matters as lead time are the least of their worries. In most cases, 'cost' means the number they get when they add up the 2/3 or so of the direct labor time they can track plus the invoice price of the materials, then multiply the labor number by about six or seven and call that the cost of a widget. That number is about as accurate as calculating the temperature by looking at a lake and deciding that, since the lake isn't frozen it must be greater than 32 degrees, and it's not boiling so it has to be less than 212, so the temperature must be exactly in the middle - 122 degrees. Precisely 122 point 0, in fact.
For a company to take a manufactured cost that includes a 40-50% wildassguess factor and compare that to an fob Guangdong price, then decide to close a factory and throw people out of work based on the comparison is somewhere between insane and criminal. Deciding to outsource because the cost just couldn't be reduced any more, when the company never had a clue what the cost was to begin with is equally absurd.
It's too bad that Jim Huntzinger and the folks driving lean accounting chose that name. They should have called it 'Honest Accounting' or 'Accurate Accounting' or 'Common Sense Accounting'.
It is not just outsourcing decisions that suffer from the illusion of accounting accuracy. Time after time I run into people who took financial statements that contain not one valid number, backed up by cost center spending reports and product cost calculations built entirely on mush, and launched kaizen events, six sigma projects or invested in automation to affect the numbers, then were stymied when it turned out that their efforts had not improved a thing. At best, they succeeded in reshuffling the reports, moving money from one hokey account to another.
It seems obvious to me that the starting point for a company that is looking for lower costs ought to be figuring out what their current costs are. It may come as a surprise to people who have never really been in a factory, but the fifth grade arithmetic upon which accounting systems are based comes up a tad short in representing the millions of variables simultaneously moving in any manufacturing operation. No cost is absolutely fixed in a perfectly flat line fashion. No cost is linearly variable on the basis of one convenient driver. No cost is absolutely direct and no cost is absolutely indirect. Assigning costs to one of those convenient buckets does not make them so.
The harsh reality is that numbers on paper are not reality - even when they have a '$' in front of them. They are, at best, a poor attempt to reflect reality. In that regard, accounting systems are a lot like MRP systems, which also attempt to create a mathematical model of manufacturing, and do a similarly poor job. One common thread throughout lean is that it is based on reality, rather than these simple minded math exercises.
A CEO would be wise to demand lean accounting statements and do a whole lot of gemba walking - get his or her feet firmly planted in reality - before giving outsourcing or any other major change a moment's consideration.