Standardization, or high agreement (on The Lean Edge)

by Jamie Flinchbaugh on 06-18-12

This post originally appeared on The Lean Edge:

The question asked is “Are work standards individual or collective?

Standardization is a very difficult topic for most people in lean. The difficulty starts with a past practice and perception that standards are something we give people to force them to do work in a way that might not even be the most productive. Because of this, the perception of standardization is often far from its intention.

Our preference is to use the words high agreement of both what and how. The reason for these words is it conveys what we believe to be more the intent of standardization. It is a belief that we will work towards a common approach. In fact, our definition of high agreement is:

“valuing a common way or process more than we value our own way.”

What that means for the framing question, who develops the standards, informs us as to our approach more than who specifically develops it. Ideally, those involved in the process would jointly develop the standard, although we should have some expectations in place around what must be accomplished by that standard. When we can make that development approach happen, it yields the best result because the high agreement that is established while developing the standard. However, that is not always practical. Imagine you were developing a standard for 1000 service agents spread out around the world. You are not going to pull them together to develop that standard. You must develop it and deploy the standard. But you still must reflect the interests and the needs of those who are using the standard: How you involve some of them, or test the solutions with them, or most importantly, have a mechanism to provide feedback when those standards are not working.

I thought Art Smalley’s article Standardized Confusion on this question was very useful and descriptive, and I encourage everyone to give it a read. It paints a picture that standardization as a whole is more than any one tool or application. It has very is different forms depending on what you were trying to accomplish. When you pull up to an intersection, there is a standard. The person with a red light stops, and the person with a green light goes. That’s not a standard work instruction, but it is a standard.

In addition to challenging how we develop standards, as well as who we truly develop them for, we must also challenge some of the narrow views about what is meant only say standard work. I just recently wrote Standard work is not a replacement for skill and knowledge, in part as a review of some of the lessons of The Checklist Manifesto.

Start with the objective, what are we trying to accomplish? Then develop the right approach to accomplish that objective.

Have an iPad or Kindle? You can get either The Hitchhiker’s Guide to Lean or A3 Problem Solving from the Kindle Store.


  • “Start with the objective, what are we trying to accomplish? Then develop the right approach to accomplish that objective.”

    That statement is GOLDEN! One of the biggest obstacles is getting people away from looking for some “recipe.” It’s like thinking is secondary to compliance.

    I was asked recently from a department manager if his Kanban system was correct. I asked him in return, “Is it working for you?”If the system is working, it’s correct. If it’s not, it isn’t regardless of the “rules.”

    Will Grimes June 18, 2012 at 9:08 pm
  • Will, thank you for sharing your comments.

    Recipes rarely work. Even a cooking recipe will end up differently in the hands of different cooks.

    To your point of “if the system is working”, the Check of PDCA is really all we need. When we stay true to that, even when implementing standards, it is hard to fail too badly.

    Jamie Flinchbaugh June 18, 2012 at 9:21 pm
  • I do like your terminology of “high agreement of both what and how” and your definition of High Agreement (aside from are vs. our ;-). We tend to use “consensus” and “can you live with it?” terminology as we develop the standard work where there was none before. Regardless, mandating a standard without at least some team member representation up front will make even the seemingly best standards hard to sustain.

    Tim Z June 19, 2012 at 11:09 am