When Metrics are the Problem

Metrics drive behavior, and the wrong metrics drive the wrong behavior.

Leanblog (now there’s a user name) writes:

At my company the direct labor productivity metric is the main decision making point for any activity at the plant. As such there is a lot of management resistance to using operators to work on lean activity. It is believed by most of the management team that lean activities will not pay off in direct labor productivity improvement. And it is true that if I use direct labor for a kaizen event the department will take a direct labor productivity hit in the month the event occurred. This is especially true if we are working on flow improvement, 5S, visual management, LTA which may never improve direct labor productivity. Over several years this culture remains an obstacle to Lean.

It’s an old-fashioned way of thinking. You want the workforce to be as productive as possible, so you measure direct labor productivity. This tells you whether each worker is producing as much as he or she possibly can.

But here’s a thought: Direct labor productivity is not a good metric. It doesn’t matter whether each worker is producing as much as possible. What matters is whether the plant is producing the amount of product the customer wants. And that is not the same thing.

Convincing management to give up their entrenched belief in direct labor productivity as a metric is not likely to happen. What’s needed is more of a stealth strategy.

Don’t talk about direct labor productivity. Talk about other metrics – or better yet, set a goal. Tell management that if they give you the people you need to make lean improvements within a specified period of time, say a month or a couple of weeks (the less time, the better), you’ll give them measurable improvements in cycle time and capacity – in other words, the ability to move more product out the door more quickly. You might be able to lay some groundwork yourself first, perhaps with a spaghetti diagram that will show wasted movement. Working on flow improvement, one of the lean approaches mentioned, sounds like a good idea.

By the way, on the Productivity Press Web site, you can download a free article (originally published in the Lean Manufacturing Advisor newsletter), about good vs. bad metrics.

 Other comments?