In their book, Improving Performance, Rummler and Brache talk about three levels of performance: the organization, the process, and the job/performer level.

At that third level, one of the ways that managers deal with problems is to try and fix what’s broken: train, transfer, coach, counsel, threaten, discipline, or–beloved by makers of allegedly tough decisions–replace them.

As they say, assuming that defective people are at the root of all performance problems is as foolish as assuming that a bad battery is at the root of all automobile malfunctions.

Rummler and Brache advocate a systems approach to on-the-job problems. The quality of the outputs (the results) is a function of all the other parts of the system: the inputs, the performers, the consequences that result from their action, and the feedback.

Some time ago, I made a chart of my own based on Rummler’s thinking:

But — what about motivation?

“People occasionally tell us that we’ve missed a factor…that the key performance variable is motivation (or desire, or drive, or attitude, or morale). We agree. However, motivation is a symptom.”

They go on to say that if…

• If (physically, mentally, emotionally) capable people…
• Who are well trained (where they need skill and knowledge)…
• Are placed in a setting with clear expectations (performance standards, well communicated, that people can actually meet)…
• Without task interference (such as too much work, inadequate time or tools, incompatible demands)…
• With reinforcing consequences (that are timely, meaningful to the performer, and support the desired performance)…
• And appropriate (relevant, accurate, timely, specific, understandable) feedback…

Then they’ll be motivated.

We have found that about 80 percent of performance improvement opportunities reside in the environment. Usually, 15 to 20 percent of the opportunities are in the Skills and Knowledge area. We have found that fewer than 1 percent of performance problems result from Individual Capacity deficiencies.

In other words, 80% of the performance-improvement potential doesn’t come from “choosing the right people,” and it certainly doesn’t come from “training them well.”

By “the environment,” Rummler and Brache mean things like job design, workflow, tools, systems, processes.  To remove task interference, restructure the job. If you don’t have specialists in this area, “a work team of incumbents, supervisors and analysts can usually, without any sophisticated technology, make the changes necessary to remove the most significant forms of Task Interference.”

Or, let the people closest to the job figure out how to remove obstacles from getting the job done.

They emphasize that you can’t do this in isolation–the goals for the job/performer level have to align with those at the process and the organizational levels.  In fact, this is why many job-improvement efforts fail: they make the job more efficient, but they don’t link it to the process goals.  The purpose of a job is to help carry out some process–so without that link, the job’s not making much sense.

That’s what Rummler and Brache call dirt in the engine of performance.  If you know how to diagnose, you’re much more likely to find it–and to find ways to remove it.

Diagnosis chart by Dave Ferguson.
Lego troubleshooting by Philocrites / Chris Walton.