Unhealthy Measures

It seems like every few months someone tries to claim performance management is bad for organizations. 

The latest one comes from Phil Dourado who writes, “Target-based systems distort everything and most people’s behaviour.” Phil bases his indictment of performance management on several stories about the UK’s National Health Service in which “brilliant, dedicated people [are] pushed into ludicrous behaviour by top-down target-setting.”

I’m not going to defend anyone’s actions but the fault isn’t performance management or even poorly set targets. Remember the classic saying ‘what gets measures, gets done’?  Well, the National Health Service used the wrong measures so the wrong things got done. Someone doesn’t understand the difference between activity measures and outcome measures.

Take the first example about ambulances with patients in the back of them waiting outside hospitals, rather than bringing the patients inside. Apparently, the central government mandated a ‘waiting time limit’ measured from when patients enter the hospital. If patients wait outside, the hospital is less likely to exceed the maximum waiting time (aka target).

By why are they measuring waiting time? One of the overall objectives for the National Health Service likely is to ensure that people are healthy. Someone who set up this measure likely assumed the less time patients waited, the more likely they could be appropriately treated, and the healthier they would be. 

The problem is they’re measuring an activity that may or may not be tied to the desired outcome. For example, a patient with a life-threatening injury should certainly have a shorter wait time than someone with a broken leg. Furthermore, how do they decide whether they should care more about the maximum wait time for any one patient or the average wait time for all patients? And if they give patients the wrong treatment (as in the second example cited by Dourado), does it really matter if they did it in a timely fashion? They are using unhealthy measures.

I may be guilty of speculating but there may be a clue in the use of the phrase ‘lean service.’ Lean six sigma and related methodologies force organizations to obsess about improving their efficiency, without careful examination of whether they are actually using the right processes. True performance management instead focuses on effectiveness – better processes for the situation, rather than improving each process.

In my opinion, this obsession with efficiency is unhealthy. Personally, I’d rather wait a little longer – at the emergency room, on hold for the call center, at the grocery story – if I got the right treatment/answer/price the first time. And if I didn’t have to go back.

The cure is simple: focus on the outcomes, not the activities. Use healthy measures.

10 Responses to Unhealthy Measures

  1. Mike R May 17, 2007 at 4:56 pm #

    C’mon, aren’t you missing the point? In emergency rooms, it’s all about life and death so time is critical. That means it’s all about efficiency. If I’m having a heart attack, I better be given treatment as quickly as possible.

    • Jonathan Becher May 18, 2007 at 1:38 pm #

      With all due respect, emergency rooms must consider effectiveness. This is what is at the heart of the triage process. Which patients *must* be treated first? In other words, not how do they maximize the number of patients treated but rather how do they maximize the impact of the treatments.

      If someone shows up at the emergency room with a non life threatening illness, they should have to wait longer than the person having the heart attack. Using average waiting time misses this entirely and causes odd behavior as was reported in the article.

      • Mike R. May 18, 2007 at 5:51 pm #

        I suppose we should agree to disagree. From my point of view, if you can’t measure it, it doesn’t exist. “What gets measured, gets done.” If we’re going to improve our performance, we have to know what’s not working and be able to predict future performance. This means we need to collect data, organize it in some kind of store, and then apply sophisticated analysis to it. Talking about objectives and outcomes isn’t going to change that reality.

  2. Robert E May 17, 2007 at 8:35 pm #

    There is a great story in “Blink” by Malcolm Gladwell about heart attacks. A hospital in Chicago was treating everyone who came into their emergency room complaining of chest pains. Chest pains can come from a variety of conditions from heart attacks to just some internal gas. Efficiently treating all the patients with chest pains as potential heart attack victims tied up a lot of resources and a lot of hospital beds in a resource-scarce hospital.
    But how could you tell whether someone really had a heart attack or just a bad case of gas? Doctors were presented with a study that showed you only needed to look at a few indicators to tell who was really having a heart attack. Most of the doctors were skeptical. Doctors felt that they needed many tests because isn’t more information better?
    When the doctors agreed to only look at a few indicators for one month, they were surprised how accurate — and how quickly– they could get patients to the correct level of treatment.
    They were able to be effective by using a few indicators that provided both better care for patients and saved scarce resources for critical patients in need of that attention.
    So even in the case of someone believing that they were having a heart attack, getting the right treatment is just as important as getting treated quickly.

  3. nmarks January 7, 2010 at 7:39 am #


    I think you have maligned Lean Six Sigma. It starts with what is valuable to the customer, then eliminates muda that is not contributing to that product/service, and optimizes the rest.

    Everything in a Lean Six Sigma world would be focused on the quality and timeliness of patient service, including what is measured and managed.

  4. Rob Allen April 30, 2014 at 3:35 am #

    A thought provoking discussion. Unfortunately the ‘what gets measured gets done’ mantra is usually interpreted in the corporate world as I need to micro manage ‘my staff’. I have seen Lean Systems Thinking stimulate creativity amongst frontline staff and if done correctly it can empower those who do the work. Conversely it can threaten the management hierarchy. It seems that the obsession with measuring things gets in the way of dealing with the real challenges that exponential change brings. The ability to cope with ambiguity, to think holistically and how to work collaboratively, sharing the best knowledge from interdisciplinary approaches seems more salient for what lies ahead rather than pursuing a whole range of meaningless Kpi’s. How about a new mantra for the next decade ‘give up command and control’ all ye who enter the ecosystem of management and work

    • Jonathan Becher May 3, 2014 at 1:18 pm #

      The “what gets measured gets done” mantra also bothers me. In my experience, it means that people tends to measure what’s easy to collect, rather than what’s important. I often write about the difference between activity and impact measures.

      My favorite performance management phrase is “not everything that can be counted counts and not everything that counts can be counted.” More on this here:


  1. Performance Anxiety « Manage By Walking Around - August 1, 2008

    […] the desired improvements. It could even lead to unintended behaviour.” Indeed.  Remember the story of the hospital that kept patients in ambulances outside the emergency room so that they could […]

  2. More Gloom Than Doom « Manage By Walking Around - September 7, 2008

    […] as I blogged in Unhealthy Measures more than a year ago, when the National Heath System mandated a waiting time limit, ambulences […]

  3. The Danger of Increased Efficiency - Manage By Walking Around - December 9, 2020

    […] I couldn’t agree more. In my own industry, I have often argued enterprise software overly emphasizes efficiency. This is evidenced by the focus on automating existing processes and can lead to unhealthy performance metrics. […]

Leave a Reply