You Get What You Measure
It is often said “what gets measured gets done.” I don't quite believe that, but measurement has a clear impact on behaviors within an organization. Most of those behaviors are unfortunately negative. You can do more harm than good with measurements and incentives. They must be carefully crafted, and also coupled to a strong culture that overrides the measures. A strong culture can determine the right behaviors, and then the measurements are only an influence, not a primary driver.
I have a silly little story that demonstrates how measurement impacts behavior, even real time. I was a young material handling supervisor in a large and aging manufacturing complex. My direct boss at the time was a CPA. Nothing against CPAs, but he certainly wasn't suited for a job managing logistics and material handling. This was a tightly packed plant with lots of material moving in and out. The result was a good deal of forklift traffic. My boss decided one day that the fork trucks were driving too fast. I'm not sure that was the case. My guess since all the employees hated this guy, that when he walked out of his office one day they gave him a little “drive by” scaring him a little, just to teach him a lesson. So he decided he needed to get people to slow down. He asked me, in a way making it clear this was expected, to set up what could only be described as a speed trap.I would put two marks on the mail aisle a certain distance apart and then time how long it would take someone to go line to line, giving me data to calculate actual speed. Everyone got a kick out of this, and had quite a bit of fun with me, especially since they knew who the idea came from. They would speed along right up to the line, and then slow down to an absolute crawl through the area, then speed up again. One person even put up a cardboard “Speed Trap Ahead” sign up. I'm sure if I really wanted speed data, I could have found another way. But knowing how pointless the exercise was, I choose the “limit my pain” path. The worst part about this is that people were speeding because they had to. There was no place to put material, the line wasn't stable enough to consume it evening, materials were rushed in the door and straight to the line – there were essentially a host of systemic issues that meant the material handling staff had two choices: rush around and speed, or let the line go down.
Too often we try to measure the wrong thing. We specifically measure the “symptom” instead of the symptom. Forklift speed was only a symptom. It wasn't the actual problem, just an indicator of the problem. Later on in my career, I was a manufacturing and process manager for part of a plant. We were measured on many things, but one indicator was the number of ideas submitted. My peer on the other side of the house had gotten a grade of ‘A' in Manufacturing 101: How to Beat the System. The result was he put every little improvement down into the system, and then corrected it. If a bolt was loose, it was submitted as an idea, fixed and responded to. He was simultaneously the most prolific generator and respondant of ideas. The next month when the report came out, his group had generated 10x the number of ideas as my group. I got the report on my desk with a frowny face drawn next to my number (no kidding!). The lesson: game the number. The number of ideas was only an indicator of a healthy organization capable of finding and implementing opportunities to improve.
I believe we make three critical errors when developing measures for our organization. First, as I already mentioned, we measure the symptom instead of the system. Any symptom can be manipulated, but if done right, not the underlying system. Second, we focus our metrics too narrowly. If you measure the individual, you will get individual behavior at the expense of the group, for example. And lastly, we forget that measurement is only one tool for evaluation, not to be used as the only tool. There are many tools and methods to evaluate the current state, including things such as observation.
What measurement error is your organization committing?
Please check out my main blog page at www.leanblog.org
The RSS feed content you are reading is copyrighted by the author, Mark Graban.
, , , on the author's copyright.
What do you think? Please scroll down (or click) to post a comment. Or please share the post with your thoughts on LinkedIn.
Don't want to miss a post or podcast? Subscribe to get notified about posts via email daily or weekly.
- Episode #11 of the “Lean Whiskey” Podcast: A King is Saved, and a Workweek Shortened - January 10, 2020
- Guest Post: Use forced habits to change behaviors - March 13, 2015
- Leading Lean A-Z: K, Be Kinetic - January 8, 2010
I assume this was a union environment, with the forktrucks. You were lucky that they WANTED to drive fast. What were their incentives? If their incentive was to finish their deliveries as fast as possible so they could get to their next break time, then I understand why they were driving so fast. Otherwise, I’d be cynical enough to not believe that they were driving fast in order to keep the line from shutting down. Most everybody is good about acting in their own self interest, if they’re not feeling like a true partner in the process through the way management is treating them.
Far too cynical – most people come to work wanting to do a good job. The system, and the environment, just don’t allow it. Most of these folks didn’t take their full break. Yeah, it was a union shop. But like most people, they too wanted to do good work.
Cynical yes, but it’s grounded in my previous reality at GM. Seeing people laying on a workbench, behind a little “fort” they created behind toolboxes, sleeping the afternoon away… hardly evidence that everybody “wants to do a good job.” But, I blame management for creating the situation, over the long term, where people don’t feel like they can contribute. Nobody comes in and wants to sleep on their first day on the job. GM management sucked the life out of people, so I was working with folks who had 30-35 years of accumulated scars from bad management. I hardly blamed the UAW guys, as individuals. They were dealing with a horribly dysfunctional system the best way they knew how.
That said, how do you measure the culture factors that lead to those bad situations? Do “employee satisfaction” surveys really work? Should managers be held accountable for how “happy” their employees are? What happens when you do a survey and you learn that 80% of employees don’t trust management? Does the CEO get fired for that? Not likely. Does anyone have good examples of metrics on Morale or Culture?
Interesting juxtaposition between Mark’s later post about AOL and this post. In Mark’s post Dan talks about managing behaviors not numbers and then Jamie talks about measuring symptoms. It seems that the obvious culture war at AOL in OKC is not limited to one company.
Managers don’t dispense happiness, but they, through their own behaviors, create a work environment that promotes involvement of their employees in getting work accomplished and gaining satisfaction from doing it well. We call it “catching them doing the right thing.” After a few “catches” a manager’s job becomes more of a Socratic leader helping his team and others to define what the right things are. Union plants are different from non-union plants due to the number of people who have a say in managing events, symptoms, etc., but the leadership of “catching them doing the right thing” is the same and it is just as powerful.
My teams always help develop their measurements because they want to see a representation of the quality of work they are doing, and their personal investment in “making their numbers” makes their focus on their work more important to them. The key is helping them become successful in their own eyes.
Good leadership is good leadership regardless of whether a union environment or not. Teach / coach / encourage / reward the right behaviors; provide the right environment for teams to grow / learn / manage; give people the tools they need to succeed.