Today’s obsession with quantifiable objectives is more about office politics than performance

(This is a guest post by John Fletcher. John is an Englishman now resident in Europe, with a long career in the public sector in several countries. He has spent a good deal of time in working environments outside the Anglo-Saxon world, and has written and lectured on organizational issues.)

I was working for a well-known European government a couple of decades ago, in the days when quantifiable objectives for performance measurement were new and exciting — at least if you were excited by quantifiable measures of performance objectives. We had an office in the department where I was working which spent most of its time involved in complicated international negotiations.

So, when it was approached by the management planners, with their Boy Scout-like enthusiasm for the task of performance measurement, the department in question said, “Sorry, no can do. You don’t understand — our work can’t be reduced to quantifiable targets, because too many other people in all sorts of countries are involved; no one organization, or even country, controls the process.”

“No,” said the Boy Scouts. “You don’t understand. This is a political directive. If you don’t have any quantifiable objectives already, you’d better make some up.”

Measuring what can be measured, not what truly matters

Eventually, after a lot of pointless and wasted effort, they did. Like most other government departments, they had a fair amount of sensitive material, and from time to time somebody would go home frazzled after a twelve-hour day and leave a document lying around. This usually meant a visit from the Gestapo (the security department) the next morning. So they decided to set themselves a quantifiable target: to reduce the number of these security violations. It was an act of desperation, completely unrelated to the work they were doing, but the Boy Scouts nevertheless went away satisfied.

I think of this story fairly often, when I read about the equivocal and frequently counter-intuitive results of performance measurement techniques.

There are a lot of reasons why they don’t work well; chiefly that — as in the example above — they don’t measure what’s really important. Instead, they encourage people to put effort into things that can be quantified, irrespective of whether they actually matter. Indeed, it’s something of a cliché that anything in an organization can be measured — except what’s genuinely important.

The psychology of facing set targets

There’s another factor which explains why performance, paradoxically, often declines when it’s measured. This one has little to do with management science, but a great deal to do with human psychology. Consider the following example.

You are the head of a department responsible for external relations, in a local or national government, an international organization, or in the private sector. One of your tasks is to answer outside inquiries. Your organization has a policy, as most do, that all queries should be answered within a specific time (let’s say two weeks) unless they are especially complicated and difficult.

This is all fine until the management planners come along. They explain that they want you to set a formal, numerical target for the year — and each time you hit that target, you will get a bonus.

Because you have never collected the statistics before — you regard it as a point of honor to answer every query within two weeks if you can — it takes some time to do an exercise which shows that you actually managed this 95% of the time last year. So is that going to be your target? Well, no, because you may not hit it next year for some reason or other. Then you don’t get a bonus. What you commit to is, say, 90%, because you know you can do that.

So you duly set that as the target and report against it. And the figure for the first year is indeed 95%. Bonus. Great!

But then you and your staff start to think, “Wait a minute, we don’t get bigger bonuses if we hit 95%, do we? So why bother to stay late on Friday to finish that reply? We’ve already hit our target for this year, and it’s only the middle of December. We could leave early and do some Christmas shopping.” And so, of course, the actual level of performance will slowly decline . . . 94%, 93%, 92% — but you still get your bonus, because the target is 90%.

Treating human beings like Pavlov’s dogs

This is why performance measurement is almost always a bad idea. It converts something which is a matter of pride and professionalism into a matter of simple rewards, usually financial. It’s treating humans as performing animals; giving them a biscuit if they get it right.

It does, of course, have several political advantages. It’s easy to understand; it fits neatly into organizational reports (”Our target of answering 90% of queries within two weeks was reached for the third successive year”); it gives under-trained and under-skilled managers the illusion of control; and it complements very well the current obsession with suspicious micro-management of organizations and staff.

If you think that these advantages are of little importance compared to the damage objectives of this type wreak, you are out of step with how organizations work these days. Anything that fits into standardized reporting, gives the illusion of tough control, and puts the blame for mistakes onto junior staff is priceless.

As I said, quantifiable objectives are really all about politics, not performance. As a result, people face a stark choice: either get with the program or suffer the consequences — even if the result of all the measurement is to make the business worse.

[ratings]

Technorati Tags: performance, performance measurement, objectives, setting objectives, numerical targets, organizational reporting, benchmarks, benchmarking, priorities, standardized reporting