six-sigma

Six Sigma is a sheer gimmick

Talks about Processes and methodology begins where innovation and creativity ends.

Six Sigma especially tries to quantify everything while qualifying nothing. And it miserably fails in any department where a little bit of innovation and creativity is needed (not just wanted). A fortune article (Jan. 22, 2001) stated that “of 58 large companies that have announced Six Sigma programs, 91 percent have trailed the S&P500 since”. That is a huge number of failure exclusively by force-fitting such redundant processes into a highly creative and technical organization.

Fanboy statements

I have noticed a very interesting pattern that is universal across the software industry. Advocates and fanboys of methodologies like Six Sigma are most likely those who does not have the skill to grasp technically and hence have taken a safe recourse by walking the “i am into management” route. They are cheap wannabes trying to make a mark doing nothing. I believe management is something inborn. And you do not need an MBA to make you a good manager. I have deep respect for someone who is inherently good in management – no, not the “I am an MBA so I am a manager” types.

Where Six Sigma fails

The whole idea of force-fitting a methodology does not just seem right for the Software industry or any such industry where creativity is a prerequisite. It may be of *some* use from a manufacturing perspective but trying to forcefully fit that into software development is utter nonsense. A detailed calculation would reveal that processes like six sigma is a colossal waste of time, effort and money in any other industry other than manufacturing. Most of the so-called “tools” are sheer common sense – you don’t need to be a sheep in believing Six Sigma brought about these things. You do not need any of those tools as long as your brain is working. Think about it, there was a problem – it was rather a silly mistake – you realized it and fixed it. Okay, you announce to the team about it and document the best practices reflecting this. This is a minimal process that has nothing to do with Six Sigma. I would say this much is more than enough. But with Six Sigma that is not enough. You need to conduct a Rapid Problem Solving session. This is where the methodology fails – in trying to make itself relevant. And then document that in some arcane format  – usually a Fish Bone diagram. Doing so not just wastes developer’s efforts, but also sucks in cost. At the end of the day, the cost spent for saving that money is much more than what was intended to be saved.

You do not need a methodology to tell you that you were wrong or your planning was wrong. If the problem is big – it would have been figured out early. But typically smaller problems are hard to find and if not plugged, it may lead to huge problems somewhere half way through the job. Say, you are constructing a bridge. You start at a point and construct it half way until you realize that there had been a small problem somewhere in the beginning was wrong. Even though it is small, you realize you cannot continue the construction until that is fixed. The bridge has to be demolished and reconstructed – a HUGE waste. Proponents of Six Sigma would argue that adopting it would have avoided this situation. Let me tell you, Software development is a different beast altogether. You type 100 thousand lines of code and suddenly find something is a little wrong somewhere. What you do ? Shift+Delete ? Nope. You open the affected source files, fix them and recompile. An star developer would take few hours for this – and work could resume as if nothing really happened. Why should time, effort and money be wasted in adopting these methodologies for software development ?

The unnecessary link with Quality

Methodologies like Six Sigma are not even remotely connected with Quality. It is a set of procedures aimed at quantifying data. Nothing more. And that mythical 99.9999998% quality is a fairy tale. By tossing out terms like “Six Sigma” and “99.9999998% quality”, people who make a living out of duping managements effectively exploit the naivety of the management and when the management realizes that too much of time, effort and money has been wasted for little or no gain, it would mostly be too late.

Methodologies like Six Sigma necessitates periodic methodology adherence audits which in turn forces the managements to appoint few “Quality Consultants” (the people who dupe the IT managements) who otherwise do not have any practical job to do at that organization. And they do not come cheap. I guess the salary for them starts at around $80000. And this team will not be contributing to the bottom line though they would claim they were critical and instrumental in company achieving its results. Claims are just claims – There would be no data again to substantiate their claims. I don’t understand how they call themselves as “Quality Consultants” as they do not even know what is the technology all about and do not really care about the quality of the deliverable. They are worried just about the adherence to some process and they believe a process can automatically code, compile, test and release a product.

Incapable manager

Such an impractical methodology is at best confined to libraries rather than the main stream. Incapable managers would most likely enforce such worthless methodologies and shove it down the throat of all the employees. These methodologies give a false sense of pride for an incapable manager who would tend to believe processes are more important than skills and hence if the product succeeds, he can take all the credit for implementing the so called “right” process whereas the obvious truth would be that the development team would quickly realize the mess that would be caused by these processes and hence would relegate it to paper while choosing common sense over such despicable processes.

The fate of compliant Organizations

Motorola “invented” and practiced Six Sigma and it would have gone down the drain if not for Google. GE adopted it and was on the verge of falling from the cliff until Obama came to the rescue. I am yet to hear a real long term success story that came about by adopting methodologies like Six Sigma. As mentioned earlier, almost 90% of the organizations were not able to bring in any value by adopting methodologies like Six Sigma.

A real conversation with a Black Belt

I have met a handful of process advocates – black belts and the likes and have interacted with them. Whenever I start to explain the harm these are causing to any non-manufacturing industry, they start this way :

Me : Do you see the harm you guys are causing due to you relying on processes such as Six Sigma ? 90% of the organizations fail with this.

Black Belt: Which portion of Six Sigma you have a problem with ? Do you know that Six Sigma is to achieve a 99.9999998% quality ?

Me : But Six Sigma does not talk about HOW to achieve that 99.9999998% quality. It just states that it intends to – on paper.

Black Belt : You have tools in Six Sigma you can utilize.

My Friend : (Googles to “download” the tool that miraculously gets you the 99.9999998% quality until he realizes those tools are just in paper).

Black Belt : May be the organization is implementing Six Sigma wrongly or the consultant was not knowledgeable enough about Six Sigma.

Me : Let me reiterate, there is NOTHING in Six Sigma that talks on “HOW”. So plain a gimmick.

Black Belt : (Furious) How could you say that ? Are you saying Six Sigma is a waste of time ? Is that why Companies are spending billions of dollars on these initiatives ?

Me : I would say so. And Companies could literally *SAVE* these billions and use it for their R&D initiatives instead. Sad that they spend billions to save few thousands.

Black Belt : (Jumping on the seat in fury – probably a survival instinct) Are you a happy being a cowboy coder ? (the most solid defense according to them)

Me : In such case, are Linus Torvalds, Bill Gates, and Steve Wozniak “cowboy coders” too ? Quality of the deliverable solely lies in the skill of the developers involved. Six Sigma does nothing to improve that skill. 

Black Belt : [… Goes on and on]

The truth

Companies like Google have no formal processes. But companies like Microsoft used to have them. Microsoft may not be foolish enough to adopt Six Sigma, but it certainly had a useless methodology that did cause 50% of the damage Six Sigma would have caused. A lot of star employees left the company disgruntled and most of them complained that it has become “Process-ridden” (My friends and Glassdoor reviews). Probably that explains why Google turns out to be much more innovative and faster than Microsoft. Only recently, Microsoft has downsized its process initiatives and has actually started to innovate as a result of “processes” being thrown out of the window.

Adoption of Process and Methodologies are INVERSELY proportional to Creativity and Innovation.

  • Pingback: Surface 2 Pro – A Do or Die situation for Microsoft | SaneNet()

  • dontknow

    Ironically the company that use to focus on processes, lean manufacturing on which books have been written about recalls millions of cars every year.

    • http://www.sanenet.com/authors/visualsan visualsan

      Yes, and they are not yet a part of F1 :) F1 is where quality and speed matters… Obviously with heavy baggage, that company can never get a spot…

  • Pingback: An MBA will not make a manager. It is all in the genes. | SaneNet()

  • Angeras

    Very insightful article…. All these processes might have done some good in 80s and 90s… but these are totally irrelevant in these times!!
    Clearly people who flock and advocate these kind of processes in this age are only time passers in the software industry!!.. I cannot comment about their ‘benefits’ in other industries… Being a software developer, its disgusting to see such “process” ppl acting like “the boss”… while their actual role is to support and help core developers..!!

    • http://www.sanenet.com/authors/visualsan visualsan

      I disagree with your last statement. The “process” people cannot really help or support a developer in anything. A good developer would do better without these people around. It is better to redeploy all these people to core development so that they can learn about the real thing instead of doing any non value added stuffs. It is irritating to see them meddle with something they dont have a clue about – i meant the product and the development cycles.

  • Juan

    This is an old post, but I cannot leave without give my opinion (otherwise why is this box here?)

    My very rough and uneducated understanding about six sigma is that it is a methodology to reduce variation, where you processes (whatever you’re measuring) have a 99.99966% of probability to lie within specs.

    Where I do see that all of the companies fail miserably is in adequately perform the proper statistical analysis in order to identify and be able to address the issue.

    That being said, you do not need to have a BB or super-plus-ultra-master certificate in six sigma, the thing that must be implemented as a philosophy is the statistical thinking and, as you pointed out, use your common sense.

    I have seen plenty of MBA that simply cannot identify a between a continuous and a discrete distribution, and even they try to analyze everything as if it were “normal” distributions (if we are to the point that they know how to analyze data).

    To use the DMAIC as a methodology is a nice thing, but let’s be honest… when you are in the line of fire and you are asked to solve a problem, the last resource that you have is time to seat your ass and follow a tedious methodology, you need to be frugal and solve as quickly and economical as possible.

    • Keshav

      True Juan… As you and the author point out – it is pure common sense packaged in exotic names and marketed it at a high cost to companies… But the problem is, companies maintain such teams paying millions just to save few thousands as the author points out… Common sense rules… The author is more biased to software industry than manufacturing… In manufacturing you could justify such nonsense… But there is no justificiation to adopt six sigma like processes in a software industry… Six Sigma in Software Industry does not make any sense at all….

  • TrumpCon2017

    I agree with this article. There is no process substitute for talent. That’s why crap like LSS fails and fails hard.