This will be helpful when you are planning an appraisal or performance review for a colleague or if you are helping someone else to do so or when you are preparing functional, departmental or even organisational goals.

Avoid the Common Traps

The first big trap is to look for what’s easy to measure and to use that. The second is similar – to choose a measure that already exists in management information of one sort or another.

Neither being easy to measure nor already being measured is a bad thing, of course, but neither characteristic guarantees that the measure will be either meaningful or useful in improving performance. These are simply wrong starting points for your thinking.

Focus on the Aspects You Most Want to Change

Look around you. Listen to what people are saying. Try to hear what they are meaning/thinking behind the words they use. How does it feel to be opertaing as we are?

List all the good, productive things you see, hear and feel. Also list those things which sap energy and hold back effectiveness. Which of these, if changed, would have the greatest positive impact on organistation performance?

Bear in mind what the individual, business unit or department can make an impact on and set priorities, selecting just a few issues (around three is good). If possible include good things that can be built on as well as things that are not working well and need to be built up.

What if These Areas were Already Great?

This is an exercise in creative imagination. Project yourself into a future (3, 6 or 12 months on) in which change has already taken place. Look around you. Listen to what people are saying, meaning and thinking. Notice how it feels now that these areas of performance “have already” improved.

Write down the tangible changes that you are aware of, including both objective (preferably numerical) measures and more subjective (though equally real) indicators that have changed in this imagined future. Aim for a list of 10-20 measures and indicators which have moved as a result of the changes which have “already” taken place.

Select the Most Useful Measures/Indicators

The most useful will be those most directly driven by the changes in performance you desire to make. It is not always the most statistically robust or objective measures that work best.

In fact the best reasons for selecting a measure typically have to do with the degree to which it is understood and can be tracked by those whose performance you are trying to improve.

Even a subjective scale (On a scale of 1 to 10, where 1 is the horrible and 10 is about as good as you could imagine, how have your interactions with X department been this week? And why?) can provide the basis for remarkable performance changes when used as part of a balanced diet of ongoing discussion and shared endeavour.

Work out How to get the Measures You’ve Identified

This is where measures that already exist do have an advantage. But let’s say that you’ve identified as most useful a measure that does not currently exist.

One example relevant to improving the use of a Learning Resources Centre was the number of visits by new users. Ask: “Who will gather that data, and how?”

Ideally the individual/group whose performance you are trying to develop can gather the data themselves, using a reasonably robust process. That way those most directly involved are measuring themselves and as a result tend to own the measure and the performance improvement process.

Consider the Measurement which is Most Relevant

The aim is to track real movement in performance without the noise of random, short-term fluctuations. The output of a production line might be measured and reported hourly, while for the employee turnover of that unit quarterly data would probably be more useful.

Back-Calculate to get a Trend

If this is a new measure, it is useful to know what good or bad looks like from an early stage.Working out what the measure/indicator would have been for the past two or three data points is good practice and provides a baseline for the next step.

Agree Targets and get Action Commitments

So far we have only thought about the scales that we will use. It’s now time to agree with those who can make a difference what level of performance they can achieve and over what timescales.

This is also a good time to talk about the kinds of actions they will take to help deliver the agreed targets. (In the Learning Resources Centre example mentioned earlier, the target of moving from a baseline of one new user per week – if we were lucky – to an average of one new user per day by year end stimulated all sorts of creative and energetic behaviour on the part of the Learning Manager responsible for the Centre.

Check your Measures for Possible Negative Outcomes

It’s possible to set measures which can be met in ways which will destroy rather than enhance organisational performance – like a production volume target which may lead to compromises in quality.

Some of the very public targets in the British NHS 1997-2007 are relevant – if controversial – examples of this problem. Consider a different measure or a complementary one to act as a counterbalance – the “Balanced Scorecard” concept.

Make Sure You Reward Improvements in Performance

This is not only  – not even primarily – about money. Reward includes attention, acknowlegement, publicity and many other non-financial matters. Just don’t expect to agree measures of performance in January, ignore them until December, but still to see marked performance improvements.

“What you measure you manage” is something of a cliche, but the truth is more like: “What you measure and pay attention to – you manage!”

This article first appeared on 

Strategic Media Partner

Do you like this article? Sign up to receive our newsletter  for HR insights delivered to your mailbox.