The Finish Line

I have been wrestling lately with the question of when a Six Sigma project should be considered “done”. From the perspective of the organization, it’s common to say that done means finished through the control phase or it’s equivalent, including process changes or other implemented solutions. From the financial perspective it’s tempting to say that done means solutions have been implemented, controls put in place, and savings documented for some pre-defined period of time.

But from the perspective of the Black Belt (or whomever is leading the project to begin with), these answers often don’t make sense. Rarely will the Black Belt have the knowledge, authority, or resources to implement and monitor process changes, or the bandwidth to stick around for a year to make sure that the dollars flow as expected. One could argue that perhaps the world would be a better place if Black Belts were truly accountable for these things, but in practice it is usually a process owner and/or finance rep who are responsible for ensuring that changes stick and dollars flow. And given what they are trained to do, we probably want Black Belts moving on to new projects anyway. Instead, we seem to be moving the finish line farther and farther away, and keeping Black Belts involved in more and more things.

For example, consider a classic Six Sigma project aimed at reducing scrap and increasing throughput on a high-volume manufacturing line. Any competent Black Belt ought to be able to do the basic work of defining and measuring, then using statistical techniques to sample and understand the process, and finally running the experiments necessary to arrive at a proposed solution. But if that solution involves re-training the operators, installing new measurement equipment, and reducing changeover times, then the Black Belt is probably the last person I’d want handling the implementation. Give me the process owner, a good trainer, a capital engineer, and an experienced SMED expert to do that. If that Black Belt is all of those things, fine. But in the other 99% of cases, I think a hand-off is by far the best path forward. Even if the Black Belt is accountable for these things on paper (as is often the case), in reality it seldom works out that way for successful projects.

Handpicked Content:   12 Angry Men (1957)

In fact, once a Black Belt arrives at a proposed solution, I’d argue that there’s usually little “Six Sigma work” left to be done. More often that not the change to be implemented involves other tools and skill sets like SMED, Lean, kaizen blitzes, cellularization, training or re-training, capital installations, etc, etc, etc. And given that’s the case, there’s almost always a better person or group than the Black Belt to get it done. Let the Black Belt move on to another opportunity where Six Sigma skills are needed. Why train people up on one thing and then devote them to another?

One answer to this problem is to try to make Black Belts experts at, well, everything. Hence the proliferation of various hyphenated and hybrid techniques (DFLSS with an extra week for change management, anyone?). And as a result, we get what seems like a never-ending curriculum creep: 40 hours online plus five weeks in class for Black Belt isn’t uncommon anymore, and both the online and in class parts seem only to be growing.

But maybe a better answer is to move the finish line forward. Perhaps the job of a Black Belt ought to be to deliver one thing and one thing only: the knowledge necessary to make the improvement. Maybe past that point there are others who will always be better at executing.

Comments 12

  1. Sue Kozlowski

    I’m in agreement with many of Andrew’s points. Once a Black Belt finishes a few projects, he/she is far too busy with current efforts to keep tabs on the previous projects’ process owners. Can anyone say, "leadership accountability?"

    One of the very first issues our Black Belts ran into was that some leaders aren’t capable change managers – perhaps not their fault, but nevertheless a barrier for a Black Belt who’s ready to move on to the next project.

    The next issue was that some folks decided that Black Belts were change management experts – on all topics and via all methodologies – and tried to get them to become involved in every project that was in progress (hundreds!). We strongly resisted that push.

    Instead, we’ve limited the BBs to Lean Six Sigma projects (with a very few exceptions). We’ve beefed up our Process Owner training – a special self-paced class required to be completed before signing on as PO for a new project; and a required "PO Commitment Agreement" that includes follow-up data collection and reporting – and we pay a lot more attention to this issue in our Lean Six Sigma training classes and certification requirements (currently at 9 days of class for GBs [15 for BBs] + two projects for certification, and holding).

    Thanks Andrew for bringing this topic up for discussion.

  2. sunila ahuja

    As per my view, Black belt is not only a statistic expert but a change manager, who facilitates the improvement and make that change a way of life by sustaining the results.

    So the responsibility does not end when we apply tools and give action points to the process owner, actually it should start with high spirits so as to get those changes implemented by facilitating and changing the culture.

    Sunila Ahuja
    Black Belt
    Bharti Airtel Ltd.

  3. Andrew Downard


    I think your approach is right on. Too often I see content added to Six Sigma Belt courses to address various gaps in an organization – lack of understanding of change management, lack of manager talent, poor project management skills, lack or discipline, low process maturity, etc, etc. I think it’s a far better thing to adedress these gaps head on. In my experience, to make belt training successful, 90% of the effort of a deployment chapion (or equivalent) has be elsewhere in the organization. If that work is done right, the belt training takes care of itself and the courses don’t get bloated. To your point, if your PO’s are not feeling accountability for projects it makes a lot more sense to address it with them rather than adding days or weeks to the BB curriculum so that BBs can do the job of a PO. Sounds crazy, but I see it happening a lot out there!



  4. Andrew Downard


    I content that the overlap between populations of statistics experts and change management experts is not as great as we pretend. It takes time and practice to learn both skill sets, and rare indeed is the person who is equally interested in (and good at) both things. One answer to this problem is to attempt to create change managers/Black Belts with the same training program. Plenty of companies are going this route. But another is to divide up the work; let Black Belts focus on statistical investigation and experimentation, and involve other experts when we need different skill sets. To me this makes more sense that trying to create a population of people who are supposedly expert at everything.


  5. Mike Carnell

    When we set up a deployment we process map the flow of ideas through implemented projects and clearly delineate who is the owner/accountable at what point:

    ideas – anyone
    Charter – Process Owner
    Financial Analysis – Benefits Capture Manager
    Assign Projects – Steering Committee
    Projects – Belts
    Implementation – Process Owner

    It is a little more complex than this. There are handoffs at the tollgates so no one can claim they didn’t know or understand. Once a belt completes a project it becomes the responsibility of the Process Owner to implement. Why would you place the Belt in charge? They were just trained in the SS Methodology so why give that up as a resource? Does a belt create more value in completing projects or implementing a solution where the Process Owner has the formal authority, responsibility and accountability for the process? Any Process Owner who can’t find it in their area of responsibility to implement an improvement should really be questioned as to their level of acceptance of the solution.

    There is a flip side to this as well. The system for completing projects should assure that the Process Owner is involved throughout the project even if it is nothing more than a sign off at a tollgate. When belts are trained in change management they need to understand the importance of getting the Process Owner involved or at least informed throughout the process.

    Just my opinion.

  6. Andrew Downard


    I think we’re saying the same thing, although admittedly you stated it much more clearly than I did!

    Despite what seems to me to be the obvious value in this approach, I still see a lot of bloated curricula out there. Topics that in my opinion have very little to do with the core skill set necessary for Black Belts are making their way into programs. I suspect this is driven by the lack of certain required skill sets elsewhere in the organization – as long as we’re training folks, why not train them to do eveything we need. And it seems to be getting worse. Of course this is just my anecdotal experience – I have no data to refer to.

    This isn’t to say there are no gray areas. For example, one could argue that teaching is not a core skill set for Six Sigma. I think it is, and that the ability to deliver modules belongs in the SS curriculum at some level, but I can see the other side of the argument. As another example, I also think it is healthy for BB product designers to follow a new product through to manufacturing and even sales in some cases, even though there are probably people better suited to the task. But the point is that these are exceptions, and shouldn’t become the rule.

    Thanks for the comment.


  7. Mike Carnell


    Actually I should have been more clear because I meant the post to support your position.

    If we take it to a purely pragmatic position – the material we used in Wave 1 Allied Signal Automotive was able to fit into a single binder. Most curriculum these days takes 5 three inch binders. The net effect on the dollars saved per project is virtually none. Is there value in the additional training material? I do not believe so.

    The job of a Belt is difficult enough with the DMAIC/Lean material. Adding things because it makes someone feel good is a waste of time and money. Being a belt is no different than any other job – you train in the basics. The learning should continue beyond certification as it should in any other job. Unfortunately it is easier to see with a belt because they tend to be more visible. Imagine how much different your work place would be if everyone in the company was required to read one book per month from a designated list of books and it had to be from outside their discipline.

    We were fortunate enough at Motorola to have leadership that understood we need to grow constantly and legislated training annually. It made a difference.

    Training belts – keep it to the minimum and never let them stop learning after certification.

    Just my opinion.

  8. Ben Jones

    Just speaking personally, I have found as a black belt that working hand in hand with the process owner during the Improve phase of the project has been very valuable, especially with Lean projects. The improvement idea may look great on paper in the analyze phase, but there are inevitable obstacles to overcome and refinements to be made in the "doing" of the improvement. Participating in this effort as opposed to "throwing the idea over the wall" has been both challenging and rewarding, especially since I am naturally a "thinker" as opposed to a "doer".

    Maybe an organization farther along in the Lean Sigma transition than mine would be more well suited to the approach you are advocating, where the belt acts more as a consultant than an actual change agent and moves rapidly through DMA and on to the next project. Thanks for the thoughtful post.


  9. Andrew Downard


    I have mixed feelings about this. While we should definitely take every opportunity possible to develop people, it’s not always the fastest way to get projects done. It comes down to what’s important to the organization long term and balancing needs in the short term – I don’t think there’s a generic right answer. Perhaps you’re right that it depends to some extent on where we are in the journey. Thanks for the comment.


  10. Mike Carnell

    Six Sigma now has a new role. It is viagra for natural "thinkers" so they can feel good that they are now "doers"? Pfizer will be elated.

    How much sense does it makes to take a person and train them in a process solving discipline and then suboptimize that skill set so that someone who is naturally a "thinker" can feel good as a "doer" and they can rationalize that because they aren’t "throwing it over a wall."

    We can now use that little sound bite of emotionally charged commentary to justify running our personal agenda. Part of being a team member is to be able to trust the TEAM to do tasks. That is why Six Sigma uses teams rather than the Rambo approach – our Ronco belts can slice, dice and make thousands of julian french fries.

    Ownership doesn’t come as much from working with a Process Owner as it does having a process owner actually commit themselves to the solution by the being a "doer." I believe that is called "walking the walk." Transfering ownership at natural transition points isn’t throwing something over a wall. It is allowing people to do their jobs.

    Nothing precludes a Belt from participating at some level in implementation. That is to the extent that it doesn’t create an issue with them doing the job they have in the system – create solutions to problems.

    Just my opinion.

  11. Andrew Downard


    As I said above, I think the right course does depend to some extent on the circumstances. In most cases I agree that you are entirely right. We should set up the roles and manage the transfers so that the best person for the job is employed at each step. You explained this well, and it was the point of my original post.

    While we all might have inclinations about what we’d like to be doing in our jobs, the organization doesn’t exist to satisfy our desires. But upon reflection I do admit that it could be helpful for Black Belts (or anyone, for that matter) to gain a hands-on understanding of what becomes of their recommendations as they are implemented. Lessons might be learned that otherwise would be missed.

    That being said, I think this is a development issue for Black Belts and should be treated as exception rather than norm. It involves (or ought to involve) making a conscious decision to sub-optimize the project and team in favor of optimizing personal development for an individual. For obvious reasons, that’s not a path we can or should take very often. And it’s certainly not something that should be built into a program. But nor is it a possibility we should ignore – developing a BB this way in the short term could conceivably make sense in the long term in a small minority of cases.



  12. Mike Carnell


    My reply was directed to the blog from Ben Jones not necessarily your comments. The way this is structured it can make it difficult to tell.

    Obviously a deployment can be structured many different ways. It needs to go into the planning and calculation of how many belts will be employed in a deployment if you are planning on them being heavily involved in the implementation. You will need more.

    We seem to be missing the point of the team and focusing on the belt. There should be a constant shift of ownership of any project as it progresses. On day 1 there is 100% ownership by the belt because the team does not exist. As you progress through a project there shoulf be a constant shift of ownership to team members particularly if you want them to own the solution. By the time you close a project there should be 100% ownership by the team after all they were selected because the problem affects them. Why is that throwing it over a wall? These are the same people that were involved since the beginning of the project – nobody threw anything – the team is still there that came up with the solution.

    I have a much bigger issue if the belt needs to be present. You now have a solutioin that is "personality dependent" and much more likely to fail when the belt leaves which they will do at some point. If the wheels are coming off sooner is better than later.

    Just my opinion.

Leave a Reply