Measuring Performance

by Neil I. Clark

A lot of stress is placed on "performance" and "results", but how do you measure these things in a fair and equitable manner?

The key to this conundrum lies in determining what results an employee is actually supposed to produce. So let’s look at two considerations:

1. How do you establish what a "valid result" is for a particular employee?
2. How do you measure the on-going performance of those results?

These two items will give you the tools which form the basis of evaluating the performance of any employee.

Establishing Valid Results

A good way to start this process is to ask yourself a series of questions:

• Why is this person being paid?
• Who benefits from what they do?
• What "falls over" if they are absent?

Sometimes this is very easy. Jobs which have an obvious "result", like in a sales or production area, can be recognised quickly. Many jobs, however, are hard to classify in this manner. Does a Secretary or Personal Assistant produce letters and diary entries, or is there more to their job than that? You have to be sure you have identified the most suitable result to be measured.

For the Personal Assistant, for instance, if you said "typed documents and diary entries", you would be partly right. Yes, they do produce these things, but are these the results they should be measured by?

Results Must be Under the Employee's Control

What if the manager being supported by the Personal Assistant does not originate any documents to be typed for a week, and does not want any appointments set? The PA’s results would appear to be zero. But is that the employee’s "fault"? No, the quantity of these things is not necessarily under the PA’s control.

If you take a Sales Person, however, can they increase their sales? Is it under their control to increase production? Yes it is. So sales figures (units, revenue, etc) are valid measures for that job.

But how about this Personal Assistant? If letters and diary entries are not necessarily under their control, what is? Well, it comes back to those questions above. In fact, the third question gives a very good indicator of their production (what happens if they are absent).

If the PA is away, what does the executive experience? All of a sudden they have to spend more time doing those things which the PA normally took care of. So, a more valid measure of results for a Personal Assistant would be "time saved for the executive".

That works. There are many actions which a PA can perform to save time for the executive they support. And a good PA will be constantly on the lookout for such things.

Measurable Results

It goes without saying, of course, that results must be measurable.

1. "Typed letters and diary entries" are measurable, but they are not necessarily under the control of the PA.
2. "Time saved for the executive" is also measurable, but this statistic is under the direct control of the PA.

Now, you also have to consider how easy it is going to be to measure the results. In the case of "time saved for the executive", it is possible that opinion could enter in. If all you do is ask the executive how much time they think was saved, what will you get? This does not invalidate the measure — you simply have to devise a better method. This can be done with a scoring system. Allocate points for each activity which directly contributes to the end result. The employee can do this best, as they know what each activity is.

Applying these principles to other jobs, where the valuable result is not immediately obvious, will help to sort this one out.

Measuring Performance

On the basis that a valid result is measurable, the next step is easy. "Measurable" means that it can be represented by a statistic — a number. It can be counted on a weekly or monthly basis. It can also be compared from one period to another, and this is the key.

Since you are now dealing with numbers, you have something which can (and should) be graphed. Now you can begin to look at trends and levels, and talk about targets and expectations.

Short Term Measurement

On a week-to-week basis, results will go up and down. A top performing employee will be interested in both the ups and the downs in their production. Ask someone why their results dropped last week and observe the way they answer.

1. The top performer will have already noticed (and analysed) the drop and will tell you what they have learned about it so as to avoid similar situations in the future.
2. The poor performer will usually blame others and come up with lame excuses. They will not learn any valuable lessons from it.

Weekly changes in results are simply short-term views of performance, giving clues to improving things in the future.

Long Term Measurement

When viewed over a longer period of time, statistical results give you an overall trend and a better feel for the general "health" of that part of the business. If the results are normally measured weekly, then "long term" means about two or three months. But you can also look at them over a year or two for the strategic view.

1. If the trend is upwards, there is healthy growth, of course.
2. If the trend is flat, you should be concerned, because nothing stays level for long. If it is not growing, it will eventually die.
3. If the trend is down, then the employee is definitely in trouble and you should find out what is going on.
4. If the long term trend flattens out at a very high level of production, you have a top performer who needs support (additional tools or staff).

With this additional help, their results will continue to grow.

Much of this is common sense, of course. The most important thing is that if you are going to make decisions based on the actual performance of certain employees, make sure you are measuring a valid result as the basis of that performance. In this way, your valuable top performers will be easy to see.

And, don’t forget to reward them for it!