Clear and Relevant Metrics


By Matt Wrye:

I have been implementing lean over the last 10 years at five different companies. Only during the last eight years and three companies was I truly learning about and implementing lean (and not just the tools). During that time, I have become very passionate about metrics.

Are they correct? Do they drive the right behaviors? Are they being looked at and used?

Not until recently did I spend some time reflecting on metric characteristics in areas that were using them successfully to drive improvement. Metrics must be both clear and relevant. It sounds so simple and easy but I have seen so many failures in not doing this.

First of all, the metrics need to be relevant to the employees that are targeted to use them. I have learned to ask the employees in an area I visit if the metric displayed mean anything to them. A vast majority of the time, the answer is ‘no'. It used to be shocking at first, but unfortunately now it isn't. This creates busy work for people maintaining the metric and adds waste to the system. More importantly, it shows underlying disconnect between the people creating the metric (supplier) and the people using the metric (customer). If the levels of management have a disconnect with the metrics, it is pretty safe to assume there are many more disconnects with in that supplier-customer relationship.

It isn't only important that the metric is relevant to the people using it, it must also be clear. It must be clearly understood, clearly connected to cascading up goals, and the goal must be clearly stated. Too often, I see examples of metrics that are very unclear to the user. An example might be having front line hourly employees using standard cost metrics. It is very unclear what the front line employee can do to affect the standard costing. This leads them to not drive improvement.

Another metric that I find very unclear is OEE (Overall Equipment Effectiveness). This is one that is used by many companies to drive improvement. I find it very unclear. It is a product of three factors: yield, uptime, and machine efficiency.

OEE = Yield x Uptime x Machine Efficiency

So, if OEE changes which one caused it to change? Or worse yet, what if OEE stayed flat but it was because uptime increase and yield decreased? Would anyone know? The equipment is still performing the same according to the OEE. In order to understand what is happening, you have to break apart the metric into the three components, so why add the over processing? Please, I have seen OEE misused so many times, but that is for another discussion.

When I have seen clear and relevant come together, I have seen greater improvements and greater buy-in.

One great example was an area manager where I work. When asked if the current metric (packages completed) meant anything to the employees they said ‘no' and ‘they don't pay any attention to it'. His uptime on a line was 2.6 hours / shift. He wanted to get more packages per shift but that metric didn't mean anything to the employees but uptime did. The area manager connected number of packages completed in hours of uptime in order to make it relevant to the employees on the floor. Then he took away all other metrics in the cell and asked them to hit 5.0 hrs of uptime per shift (he also gave them some problem solving help).

Within the first week the cell was hitting hours of uptime per shift. Three months later he had to increase the goal to 6 hours per shift. The area manager has been able to reduce the number of cells needed by 50% and eliminate the need for temporary employees in his area. When an employee is asked what is their goal they clearly state, “Six hours of uptime per shift.” The connection between uptime and all the other management goals is clear as everything else has moved in the proper direction, cost, delivery, OEE, etc…..

So whenever you are looking at metrics, make sure they are clear and relevant and I will bet that once that connection is made you will be very successful.

Matt Wrye graduated from Purdue in '99 with a BSIE. After implementing Lean in various industries, Matt now works for Hallmark cards as part of the central lean implementation team working to build lean systems and process excellence throughout the company.

What do you think? Please scroll down (or click) to post a comment. Or please share the post with your thoughts on LinkedIn – and follow me or connect with me there.

Did you like this post? Make sure you don't miss a post or podcast — Subscribe to get notified about posts via email daily or weekly.

Check out my latest book, The Mistakes That Make Us: Cultivating a Culture of Learning and Innovation:

Get New Posts Sent To You

Select list(s):
Previous articleThe Power of Explaining Why – A Funny Example
Next articleBook Review: "Kaizen Express"
Mark Graban
Mark Graban is an internationally-recognized consultant, author, and professional speaker, and podcaster with experience in healthcare, manufacturing, and startups. Mark's new book is The Mistakes That Make Us: Cultivating a Culture of Learning and Innovation. He is also the author of Measures of Success: React Less, Lead Better, Improve More, the Shingo Award-winning books Lean Hospitals and Healthcare Kaizen, and the anthology Practicing Lean. Mark is also a Senior Advisor to the technology company KaiNexus.


  1. Six sigma can be defined as the process which increases the customer satisfaction & in turn increases the profits to the company.

  2. When trying to determine the validity of a metric it is important to focus on the business problem. Some metrics, such as OEE, tend to be implemented because it is the "lean" thing to do, regardless of the organization's business model or situation. It can be extremely frustrating to have metrics "forced" upon a team and then have that team have to report out on them.

    In addition, it is important to keep the number of metrics to a minimum. This allows the typically limited resources to focus on the issues and corrective actions. Typically an organization can manage between 7-10 metrics. I was involved in an organization that believed that each department should have their own metrics and that all of the data should be shared at one daily "flash" meeting. While it was informative, the amount of information that was not acted on was overwhelming.

    Lastly, if possible, the metric owners, i.e. operators, machinists, etc, should be involved in developing the metric. They are closest to the process and will be able to create metrics that are meaningful to them. This requires that management trusts it people to understand the business and develop appropriate methods to track progress.

  3. I don't think OEE would be a good metric to drive change because it is kind of an abstract concept to a lot of people in most organizations. If you think about policy deployment — start with some very high level 5 yr goals and then drill down at the next level of the organization (agreeing on what should be measured and HOW we are going to get the desired improvement) — the OEE is often something that gets "drilled through". It isn't a bad trailing metric that should show year over year improvement under the assumption that you are adjusting the divisor of the performance metric to match TAKT (synchronizing to demand). So I think it is a good trailing metric but not a good actionable metric. Policy deployment needs to go 'through it.'

  4. I couldn't agree more with you, Pete and Bruce. OEE is a very abstract measurement to me and in most cases is just used because that what people are told to use. I have found that operators can not act on OEE but when you break out the components of OEE, then the operators understand those metrics and can act on them.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.