How do you describe a process to a team? There are lots of tools in the toolkit including value stream mapping, functional swim lanes, context diagrams and SIPOC to name a few. But I find they can be “a little cold” for a non-technical or cross-functional team and I want to “bring it to life”.

Take a simple example from the area I work in, general insurance. To describe the customer claims process in any depth takes time. So when you have a cross-functional team covering front, middle and back office it can mean (even with the tightest time keeping and agenda) that the people at the end of the process don’t get a look in as most of the day has been spent at the beginning/middle of the process, where all the customer interaction happens. So you end-up leaving people out.

As the title suggests my recommendation is Business Scenarios. Rather than cover the generic customer claims process, cover it in a series of business scenarios like

• The policy holder’s vehicle collided with a lamppost, no one else was involved, it happened at 11:45pm
• The policy holder’s vehicle was hit by a 3rd party vehicle from behind, both vehicles were drivable, no one appears injured at the scene, 3rd party was insured and traceable
• The policy holder’s vehicle collided was a 3rd party vehicle on a narrow street, liability is unclear, 3rd party injured and vehicle undriveable

If you start with the simplest scenario where nothing goes wrong (sunny-day) then you can rapidly walk the whole process. You can add complexity as you need to. Maybe include failures in the process and known exceptions (rainy-day) e.g. 3rd party had no insurance.

This approach really opens up the discussion as you are talking to people in the language they relate to. You get to see the true degree of variation required of the process which allows for more robust solutions.

Organisation in chaos? Emergencies erupting? Been blind-sided by the unexpected? Project a few years late and still does not work? Need to get things under control? Make way for the Corporate Fire Fighter. (Phew!)

This trusted pair of hands hits the ground running….makes rapid assessment of situation….. takes urgent action …... reports an outstanding success …….moves onto the next big fire.

You could be thinking, “Hey that’s me!”. Agreed fire-fighting can be fun, exhilarating and very rewarding for those involved. Your organisation may place a high degree of recognition and reward on people with these skills.

But is this a measure of a healthy and successful organisation?

Would an alternative model be of a highly organised machine where everything fits strategically together; risks are identified and addressed early; projects invariably deliver on time, cost & quality; business metrics provide robust leading indicators. Achieving that level of capability is difficult, very difficult.

An organisation may not have this level maturity for any number of reasons. They may be a business start-up and just about managing to keep a lid on issues as the business grows. They may be working in a highly innovative sector where new products and competitors frequently appear to “eat your lunch”.

But what about the fire starters? The leaders who raise the alarm? Is this the right thing to do?

I am no expert on management leadership & behaviour theory. It might be just the right thing to do to keep people on their toes? Creating a crisis can be a good way to drive things forward. Or is it a reactive and costly approach?

Ultimately I think it comes down to looking at the root-cause and fixing what/who caused the crisis in the first place rather than heroic fire-fighting.

Over the last 5 years I have invested considerable time & energy in trying to become a skilled continuous improvement practitioner. I am a strong believer in continual learning via direct deployment experience.

Over this same period, continuous improvement has become a main-stream product. Any business without an Operational Excellence, Process Improvement, Process Excellence, Continual Improvement (and so on) capability is now way behind the curve. This dramatic growth has brought a large increase in the number of CI professionals.

Here is the point; these days I have conversations with other CI professionals that make me wonder I have learnt the right things, things like:

• The scientific way of experimenting is to change one factor at a time
• The most important factor in sampling is population size
• Reducing overall process time has nothing to do with becoming Lean

Where did I go wrong? Case in point is the ex-GE Master Black Belt. Now I don’t like to generalise and easily the best MBB I have ever met was from GE, but my other experiences have not been so good. Ever had to explain proportion tests and what a chi-square test does to a ex-GE MBB?

How to conclude? Maybe we could agree a single version of the truth and certify against this? Maybe it’s just part of the evolution of CI? I just don’t know. But I feel the CI world changing.

In our BB training we use the terms Divergent & Convergent thinking during the Improve phase. We cover a raft of brainstorming & lateral thinking techniques to encourage people’s divergent thinking. So please take a few moments to answer this question:

• How many uses can you think of for a Brick?

How many did you get? Maybe you got just the one, “build a wall”? Or maybe you freely came-up with half a dozen? We could possibly use brainstorming here; try answering the question as if you were Pablo Picasso. Does that work?

Looking in more detail at this took me to Liam Hudson who devised this simple test to illustrate people’s thinking styles. But what Liam did was look deeper into the way we are educated to be Convergent thinkers.

The school system is based on achievement of exam results. This means being able to understand information and produce Model Answers that most accurately match what the examiner wants to see. Being good at this convergent thinking brings its rewards, recognition and good jobs become available.

Equally to what degree is divergent thinking encouraged. Starting an exam paper with, "I think the real question to answer here is......" or “I have looked at the course curriculum and believe it should be changed here and here”.

So what is the potential impact of focussing on Convergent thinking without balancing Divergent thinking? To what degree do the most successful people across industry focus on having a sense of imagination to challenge an approach? Who were the people who looked at the risks building in the financial system and saw the consequences?

It seems we should be regularly training and rewarding people for Divergent thinking rather than having as a small part of a BB training they might attend.

Picking-up on Sue’s recent Home blog, I’d like to talk about my recent experience at home.

Over this weekend my wife and I had “words” about the work I do helping on the home chores. There were a number of areas such as cooking, washing dishes, ironing, cleaning toilets, shopping, washing clothes, making beds, tidying-up, planning meals, and so on. I had no idea of the number of NVA factories at work and being a strong believer in Six Sigma I committed to resolve this problem.

I dedicated my Saturday evening and produced what I believe to be a very polished piece of work. I reviewed the key processes and created a core set of current-state value stream maps. For each of these I developed some slick data collections sheets to baseline current performance. I even identified some time saving quick wins. I shared my work and must say I was most surprised by the reaction and being told exactly where to stick my data collection sheets.

But I am a committed practitioner and realised I may have misunderstood the problem statement and goals. It seemed helping to do the chores was more important than improving current performance? So Sunday night I cooked the evening meal and over dinner suggested we discuss our differences. Luckily to support this I had previously produced a fishbone diagram and recommended a rapid brainstorming exercise followed by constraint busting 5-whys to get quick results…….

Back at work on Monday morning, I am having a tough job trying to explain my black eye. I’m sleeping in the spare bedroom and the kids think I am an idiot.

Six Sigma really sucks!

I thought I would share my views on a question frequently posed by newly trained belts. I imagine you to may have encountered this situation. I do not have a clear answer but have come-up with a theory. Could be right, could be wrong.

We talk about the discrete sampling equation used to calculate minimum sample size

Minimum Sample Size = Square of (1.96 / Precision) * Est. Proportion * (1 – Est. Proportion)

For example, what is the sample size required to find, within 5%, the number of people who are left-handed using a starting assumption that 10% of the population are left-handed?

Minimum Sample Size = Square of (1.96 / 0.05) * 0.1 * (1 – 0.1) = 138

But what people ask is, what if there are more than two categorise? What if you want to know the sample size required to find the proportion of calls split into:

• New Business Quotes
• Renewal Quotes
• Change of Service Quotes
• Administrative

Now I haven’t been able to find much of an answer to this question. I have come-up with a theory but I do not think it is statistically robust. Interested on comments and if there is an off-the-shelf statistical solution I have missed and can apply:

• Build an exploratory sample
  - Start by assuming each category is equally weighted. So the estimated proportion for each is 25%. Using quite a wide precision (e.g. 10%) you get the sample size of 72.
  - To allow for the extra categorise multiple by the total number and divide by two, hence 72 * 4 / 2, to give a final sample size of 144. The result gives you a “feel for the proportions” but is by no means accurate.
• Develop the proportions
  - You now have a feel for the proportions e.g. 60%, 30%, 10%, & 10%.
  - Because sampling theory says that 50% proportions require the highest sample size use the proportion nearest to 50%. In this case the 60% one.
  - Calculate your sample size based on 60% so using 5% precision you get 369
  - Including the extra factor to allow for multiple categorise you get 369 * 4 /2, to give a final sample size of 738
  - You can then find your confidence interval from the results obtains

I have made-up this approach and have no idea if it will stand-up to scrutiny. Hopefully I am on the right-track.

You never know it might become true like 1.5 sigma-shift……

In my last blog, Targets, I covered the situation of hitting time targets in a services environment. I thought I was onto something but wasn’t sure just what……..

Just to recap, people are targeted on delivering work within a certain time frame e.g. reply to a customer letter within X number of days. There was an understanding that this “drives the wrong behaviours” but no clarity on what to do. Here is what I have come to.

What I did was look at the current measure and split time into value and non-value as shown.

Then I looked at the two time components and put together two principles:

1. The time an item is waiting in a queue should not be the handler’s problem; it’s a management problem. The idea of pushing people to “work harder” because of variation in demand doesn’t work and alternative solutions are required.
2. The time that a handler is working on an item should not be driven by a time target but should be against a quality target. We want our people to do a great job as quality drives down rework, defects and costs.

Now the tricky bit, how to translate these principles into actual measures. What I looked for were rules for defining the measures and came to these from Vanguard

• The measure should help in understanding and improving performance – capability measures rather than targets
• The measure must relate to purpose – measure what is important to the customer
• The measure must be integrated with work – the measures must be in the hands of the people who do the work

Looking at these and my principles I came-up with two measures:

• Lead Time - The time from receipt to when work starts. Leadership and not handlers own this. They are responsible for improving Lead Times by changing the system not “cracking the whip”
• Right First Time – This is not a time measure but a quality measure. For each step in the process, the “Must-do” & “Optional” requirements are defined. These are the items that ensure the work is done correctly without defects being created. It is as simple as a check-list to ensure work is done as required.

The approach extends across the whole life-cycle. This splits the process into value and non-value adding steps. It focuses the right people on doing the right things – leadership to reduce Lead Time by reducing waste in the process – handlers to improve Quality by defining, doing and measuring what is required.

Sound good? I am already getting challenges and resistance. Would appreciate comment on the logic and how I could make the proposal even better…..

Targets appear in all shapes & sizes. Sometimes seen as positive, “we operate a target-driven culture” and sometimes negative, “targets drive the wrong behaviour”. So what is true? Given the sheer diversity of targets, I want to focus on a specific area, daily work targets in a services environment. Let’s look at a scenario.

Imagine an operator works in a services business. Work comes in three types and timing tests show each type can be completed within 20 minutes in most cases. Now imagine the operator being given items of work and being asked to work under two different management controls:

• Control 1, Work items are targetted to be completed within 22 minutes.
• Control 2, There are no targets and work items must be completed regardless of the time required

Statistically speaking, an assessment of the two approaches could be made, something like:

Ho = There is no difference between the time taken to complete work items under control 1 or control 2

I am looking at running some tests to see if there is a difference as this is related to a project I am working. But what is your gut feel on the expected performance difference?

I have tried this in a very small trial and found that when working under a time target, you focus on the time target. As the pressure builds on any individual work item because you are watching the clock you find it more difficult to focus on the task in hand and end up missing the target. You lose valuable time because of the target.

So what does this show? Does this describe an example of why targets drive the wrong behaviour? Does it show that getting it right first time saves money? Does this show operator’s pulling work? Does this show a difference between batch and continuous flow?

I’m not sure but I feel I am looking at something quite important here, just not sure exactly what it is yet……..

Happy New Year! Today, rather than talk about something deeply insightful I thought I would share one of the tools I like to use.

Ever had to arrange a project meeting? Did you need to get say 10 people’s diaries aligned, usually at short notice? Was it fun?

There are many ways to approach this simple but sometimes frustrating task.

• You could swap lots of e-mail with everyone till you get agreement
• You could force the date & location and that’s it
• You could ask someone else to do it for you
• You could forward plan and have a meeting schedule defined well in advance
• Your calendar software might even find the best date for you

Here is one way I find useful. There are a number of free on-line survey site on the internet. I take a couple minutes to set-up a survey and send it out as shown below.

Then sit back and wait for the responses.

Very useful tool.

This is just for fun and to win a real prize you need to have been very good all year and have a generous expense account.

Good old Santa wants to give you a Christmas present. He’s checked if you have been naughty or nice and the news is not good, so you have to pick the right present. One of the presents holds the all expenses-paid trip to iSixSigma Live (Miami - January 13-16) the other two presents hold copies of that classic text Statistical Analysis of Cointegration Vectors.

But Santa is a kind fellow and knowing you are a statistician decides to give you a choice. He asks which present you have picked, maybe it was B. He looks thoughtfully and tells you the prize is definitely not present A, would you like to change your mind? What will you do? Click the link to see how you do.

• No thanks Santa, I’ll stick with what I picked
• Thank you Santa, I’ll change my mind

Interested in the stats? Take a look at this: Monty Hall.

Wishing you all a Merry Christmas and a Happy New Year!