This post (*link amended to a Wayback Machine archive) – and its assenting commenters – illustrates why Agile is screwed. For as long as folks are dumb enough (ignorant of the facts) or fraudulent enough (in possession of the facts but choosing to ignore or suppress them) to believe that we can improve organisations without regard to the issue of local optima, Agile implementations will continue to fail by a ratio of 3:1 or greater (*link amended to a Wayback Machine archive).
So, I’m kind tired of dumbsters and fraudsters whining that the issue of local optima doesn’t matter.
Here’s just one note from Ackoff on the subject: “70% of Business Improvement Programs (TQM, Downsizing, Benchmarking) Decrease Performance“. If you think the same issues do not apply to Agile, then you’re dumb – or in denial, which is much the same thing in my book. And if you think they do apply to Agile, but you’re not going to mention it to your customers and clients because it’s, ahem, inconvenient, then you’re a fraud. Or worse.
If you’re looking for a non-ranty, reasoned explanation of the subject, may I refer you Ackoff, Senge, or most relevant I think, Goldratt (Theory of Constraints)?
But perhaps despite the opening, the aforementioned post (re: Bitching about local optimisations) was more about making a start, doing what you can, and gaining data from experimentation. Well, I’m all for that, with the important caveat (irony of medical analogy not lost here) of doing no harm.
And if you’re hoping (for indeed it is much more of a hope that a certainty) that “making the true bottleneck apparent” to your paying customer or client will have a positive effect, then maybe you might like to exhibit the integrity, courtesy (and courage) to place the risks and possible outcomes (scenarios) on the table before deciding whether to play the cards you’ve been dealt – or to fold?
I think Argyris would approve of that.
The Goal ~ Eliyahu M Goldratt
Agile Coaching – Maybe All You Can Do Is Send a Hallmark Card ~ Eric Laramée