Business Metrics and Performance Reviews
As part of the Business Performance Excellence (BPE) process, we help clients create and align metrics from the top of the organization on down to the individual contributor. These metrics can then be used to lead via a strategic plan and manage the day to day operations.
I often get asked, “Can we use these metrics in performance evaluation?”
The best answer is: don’t do performance evaluations. Really. There is little if any data to show that the purported value of doing them outweighs the massive expense in time, money, and anxiety that they produce. As Deming warned in Out of the Crisis:
Evaluation of performance, merit rating, or annual review… The idea of a merit rating is alluring. The sound of the words captivates the imagination: pay for what you get; get what you pay for; motivate people to do their best, for their own good. The effect is exactly the opposite of what the words promise. (Deming, Out of the Crisis, 1986)
If you don’t believe him, then you should read Abolishing Performance Appraisals (Coens & Jenkins, 2002) for a good collection of the reasons why you should abolish them and what to do instead. (Deming said that the alternative to performance evaluations is to manage!)
Still, many people out there don’t have a choice about doing performance evaluations because of their organization’s policy, so until this changes, we often have to deal with the fallout that they cause and try to minimize the damage.
Which brings us back to management and process metrics. One major factor I run into in implementing these metrics is the fear that they will be used against the employees themselves. Performance evaluations have trained our employees to fear measures of the processes whereby we create value… If I believed in conspiracy theories…
Anyway, the purpose of these measures at all levels of the organization is so that management can best understand how to support the workers who are creating the value. These measures are a result of how the systems that management has provided are working within and across the company and its suppliers and customers. In order to manage a ship and pack area, someone might well have a metric like “Number of Errors per 1000 Orders.” It might seem that such a metric should show up on that person’s performance evaluation. After all, every error costs the company money and this person is supposed to be managing this area, right?
To paraphrase John Donne, “No person is an island.” What if the manager in the ship and pack area is given faulty information on the packing slip from Sales? What if the process of packing and shipping needs maintenance that has been deferred for 6 months? And all this is often distilled down to a single “performance” number. Can a single number (or any combination of numbers) really tell the whole story of a person’s performance for the year?
Now you might respond and say, “Well of course in such a case, I would talk to the manager and understand what is going on.” To which I would reply, “Good, THAT is management!” You don’t need it on a performance evaluation to do that. What I have seen all too often though is that the performance evaluation has a number on it, and if you don’t meet that number, you lose whatever ranking or points that evaluation uses.
It is even worse for a front-line worker. If the manager is to be held to account for errors in packing, what control does the one doing the packing have? What if we linked profit sharing to the number of errors in packing? The individual worker is punished and rewarded for decisions that are made well outside of their purview. If you raised a puppy that way, you would end up with a psychotic dog.
These metrics should not be used to rate or rank employees. Doing so perverts them irrevocably. They should be used to ask and answer questions about how management needs to deploy the resources at their command to make these processes run better. The manager needs the information about errors in packing for really good business reasons. But the manager reacts to these numbers by searching for the source cause for these errors in the process, not in the people. Deming used to say that it was the decisions made in the boardroom that controlled how the process worked, and that the best a worker could do was run that process the best they could.
Yes, there are bad workers, but in my experience, 90–95% of workers want to do a good job. If you find that a higher percentage of your workers can’t get the job done, my advice is to look to the process metrics and begin to understand why the process that you have provided is failing them.
Let’s rebuild our employees trust in metrics as process measures instead.