Thursday, July 07, 2005

Six Sigma Training for the G-8?

Those who have worked for a government contractor, telecom company, or large software company in the last decade are probably familiar with the Six Sigma quality improvement program. (Actually, if you've worked for a government contractor, you've probably lived through a number of quality management trends, but this is the most respectable one I've encountered.) Six Sigma is a very pragmatic approach to management, focused on objectives and measurement. You start by identifying the results you'd like to achieve, and how those results are measured. Measurement is crucial. Managing any kind of a program without the right metrics is not unlike flying an airplane without any cockpit instruments. (Yet it's appalling how many corporate projects and government programs are run that way.) One of the common mistakes made is to confuse process metrics with result metrics. For example, a manager might measure how many hours people are actually working (a "process" metric), but that may or may not have a strong correlation to how much and how well work-product is being produced. It's important to measure the results metrics, and also to discover the process metrics that really count, the ones that do strongly correlate to results.

While Six Sigma has been successfully employed at companies such as Motorola and GE, there's no reason the methodology can't be applied to government. (In fact, the City of Fort Wayne, Indiana took Six Sigma to heart, apparently to good effect.) One project definitely in need of Six Sigma is that of aid to Africa being discussed at the G-8 Summit this week. UK Prime Minister Tony Blair is to be commended for raising the priority on African aid, but with all due respect, his "targeting" of 0.7% GDP could benefit from Six Sigma training. That's a classic mistake of focusing on a "process metric" with a weak correlation to the actual objectives. The objectives are already agreed upon: the UN's Millenium Development Goals include reducing the proportion of people suffering from hunger and living in extreme poverty by half by 2015. Those are specific measurable target results. The good Prime Minister is proceding as if there were a demonstrable direct correlation between the percentage of GDP contributed by G-8 nations in foreign aid, and the achievement of reduced poverty and hunger. Unfortunately, no such correlation has been demonstrated (and in fact, some have asserted that there is a negative correlation). Before we go throwing bigger buckets of money into a seemingly bottomless pit, we should do some measurement and analysis to get an understanding about what works and what doesn't work in achieving our objectives. It's noble for the G-8 nations to be upping their charitable contributions, but such contributions should not be given outright. Rather, they should be put into a trust fund, and released only to demonstrably successful programs, and only insofar as measurable improvement is shown. (Actually, I thought we already had a trust fund like that. Isn't that what the World Bank is?)


My blogging colleague KipEsquire is often pointing out the "Politics of the Warm Fuzzy Feeling", in which politicians want to be seen taking action, with little regard to whether it is actually effective. Sometimes they do this because they make the common and honest mistake of confusing a process metric for a result when it has little correlation. Other times they may do this cynically, because perceptions are more important to them than results. (The latter suggests the need for a Six Sigma re-engineering of what incentivizes politicians.) While I believe his motives are honest, Prime Minister Blair's call for a 0.7% GDP contribution target is definitely a misguided process metric, merely the Politics of the Warm Fuzzy Feeling. If the 0.7% GDP contribution target were achieved, you can be sure Blair will be declaring victory, and lots of people will feel good about it, while forgetting to notice whether poverty and hunger are actually reduced. I'd like to see a small portion of the aid money earmarked for sending all of the G-8 leaders to a Six Sigma training session, so that we can cultivate the Politics of Results.

3 comments:

curiouscat said...

Interestingly Prime minstrel Blair recently showed an understanding of systems thinking (specifically how targets can result in worse performance, when targets result in distortion of the system rather than improvement). In the targets in health care case the easy "politics of the warm fuzzy feeling" would be to declare victory in setting targets. In that case PM Blair realized that while the data might look better the actual results might not necessarily be better (when the system is distorted or the data is distorted).

Measures are a proxy for the actual situation and far too often people forget the proxy nature of data in process improvement.

I would agree the target of a .7% of GDP as a aid goal is an activity/input measure. Often (though to a lesser extent today than 10 years ago) the amount of money being spent is used as the measure of how much is being done. Measures of outcomes are much more important.

For more comments see: Millennium Development Goals

tom jobseeker said...

Six Sigma is first and foremost a process. We only apply tools when they meet our objectives.
Many tools will be taught, and it will be up to the Green Belt along with the team to decide when and where to apply the Lean Six Sigma tools.
It is the goal of this training program to educate the students on the application of the Six Sigma Improvement Process and the intelligent, strategic use of the Six Sigma tools.

Anonymous said...

It was a awe-inspiring post and it has a significant meaning and thanks for sharing the information.Would love to read your next post too......
Thanks:
Six Sigma Online