Simple Six Sigma: a response 17 Sep 2010Posted by Tony Law in Impact of IT, Managing IT.
add a comment
Catching up on return from holiday (including train trips in Canada with power and free capable WiFi at every seat … then back to the UK, what happened to the wired traveller??) I found Six Sigma tools for the project manager’s toolbox, published in Tech Republic a week or so ago by Patrick Gray.
He makes a good valid point that it’s easy to get overpowered by Six Sigma, and perhaps that’s why it’s fallen somewhat out of favour (I paraphrase). And he goes on to outline his take on the worthwhile things that a project manager can take from the panoply of Six Sigma tools.
Well, I trained in Six Sigma too when the IT organisation I was part of embraced it wholeheartedly. And if you haven’t figured it already from some of these postings, I’m something of a sceptic about high powered tools. I also know a little about statistics – not a great deal, but enough to notice where Six Sigma is pragmatic rather than rigorous in its statistical techniques.
So I went into the training course unconvinced of its value. I wasn’t doing the sort of thing that measurement and process improvement techniques would apply to, surely? But I came out a convert (though I never did the exam – I didn’t need to be a Green Belt). Sure, the really heavy process stuff needs someone with deep understanding and lots of practical experience. But for the ordinary working manager, I found it has real value. And it really did help me measure and improve my service: the acid test.
So: Sig Sigma tools for project managers? There’s surely a fundamental misunderstanding right there in the title. Six Sigma is not about project management. It’s about designing a process improvement, which of course may be implemented by a project, but Six Sigma is about analysing and specifying the change, and about measuring “before” and “after”. It’s not about “during”.
Not surprising then that two of the three tools he recommends for project managers are (as he himself points out) not Six Sigma tools specifically. Stage gates are pure project management. Process mapping is not specific to Six Sigma. Though, rightly, he points out that Six Sigma does have some excellent insights into process mapping. His third point does go to the heart of Six Sigma. The whole of Six Sigma is about planning and specifying an improvement, and then being able to demonstrate that an improvement has been achieved.
Here are my take-aways from Six Sigma.
- Baseline first. If you don’t know what you’ve got, how will you know if you’ve improved?
- It takes eight measurement points to establish a trend, whether it’s the baseline data before you touch anything, the indication that something’s amiss, or the after-the-fact check that proves you’ve achieved your objective
- There is natural variation in any process. Process improvement may be about reducing variation as well as (or instead of) changing the mean value of your measure, but don’t even talk in terms of zero variation
- “Don’t tinker” – the eight points rule stops you over-reacting to an occasional outlying point. Statistics tells you there always will be outliers
- In your analysis, listen to the “voice of the customer” and the “voice of the process” – in other words, do rigorous and honest analysis of the current state, including where the users are and are not happy. Look for root causes, not symptoms
- Do one thing at once: baseline, analyse, act, check. You need another eight points at the end to establish if the process has actually improved
- You can’t always directly measure what you really want to know. You may need a good surrogate measure, which in my case was to count repeat business as a surrogate for the customer’s perception of value
- When your measure tells you you’ve stabilised an improved process (i.e. the new behaviour is now embedded), stop measuring it. Go on to the next improvement, which will be driven by a new metric
I haven’t kept my Six Sigma knowledge current in the last two or three years. So the fact that this much has stuck must prove something! Six Sigma, for me, was an amplification of what Jan Leschly (a former high-ranked tennis professional) used to say when he was Chief Executive of SmithKline Beecham. If you’re not keeping score, you’re only practising.
• Six Sigma tools for the project manager’s toolbox, Tech Republic IT Leadership, Patrick Gray, 10 Sept 2010
• iSixSigma (for general information, or just Google)
• My Six Sigma trainer was John Nugent of JRN Consulting