Myth: If You Can’t Measure It, You Can’t Manage It

Posted In: ,

It is wrong to suppose that if you can’t measure it, you can’t manage it – a costly myth.”

– W. Edwards Deming, The New Economics.

One of the quotes you will see quite frequently “incorrectly” attributed to Dr. Deming is, “If you can’t measure it, you can’t manage it.” I suppose you could say it is technically correctly attributed to him; after all, it is a direct quote taken from the quote listed above. However, I think it would be more accurate to say it is misattributed to him.

If you read the original quote carefully, it is saying the opposite of what most people think when they read the abbreviated quote. This is one of the dangers in using quotes without context. For that reason, we try to provide context for quotes by W. Edwards Deming in our quote database.

Why is the shorter quote used so often and (mis)attributed to Dr. Deming? My guess is that he did stress the importance of using data to confirm beliefs about which management strategies and practices are working and which are not. And he did so much earlier than it became commonplace to see things this way in relation to managing organizations. In addition, several of the tools he recommended involved using data (Plan-Do-Study-Act improvement cycle, control charts, etc.). And often when people are told about Deming they get a very short introduction, which leaves out most of what he said, and they get the idea he was just a statistician.

So when people see a quote emphasizing the importance of data attributed to W. Edwards Deming, it seems sensible that he said it. And it is likely shared so often because people notice that their organization is flailing away when they would benefit from using data to improve their management of the situation. So the quote appeals to people who think that their organization fails to use data when they should be using it.

Dr. Deming did very much believe in the value of using data to help improve the management of the organization. But he also knew that just measuring things and looking at data wasn’t close to enough. There are many things that cannot be measured and still must be managed. And there are many things that cannot be measured and managers must still make decisions about.

I wrote a post on my Curious Cat Management Improvement blog about how to manage what you can’t measure (in 2010).

Using data to evaluate what is working and what isn’t is a very valuable management practice. And it is still a practice that is used far too little (even though it is used much more than it was 30 or 50 years ago). But much more than managing what you can measure is needed to manage organizations well.

Related: Unknown and Unknowable DataVideo of Dr. W. Edwards Deming: Deadly Diseases of Western Management (one of which is “Use of visible figures only”)

8 thoughts on “Myth: If You Can’t Measure It, You Can’t Manage It”

  1. My management style dictates the need for objective and subjective data. I can measure the objective data yet find it difficult to measure subjective data.
    Of course, the hunt for leading indicators consist of both objective and subjective data. I find it easier to understand objective data and the math associated with Lean Six Sigma is a very good example of that.
    Another good example lies with cost data. Whereas Actual cost data had better be objective and measurable, the explanation of cost variance associated with the Actual cost is sometimes subjective. In addition, in my opinion/experience calibrating costing models is subjective making the budgeted cost estimates subjective.
    Albeit I have read about Deming and admire his work et al, I’ve never met Deming nor would I know how he would explicitly think.

  2. It is fundamentally correct that managers deal with situations every day that they must manage without any measured data. This is the nature of increasing complexity in business. We say they do this via ‘judgement’ or by ‘applying experience’. This is just a subconsciously assessed action based on what the manager has seen or experienced before. What is really at issue is the idea that once that complexity is managed it should continue to be managed in the same way every time.

    Traditionally we allow managers to deal with situations and respond without formal measurement as we are not certain how frequently this situation will occur. If it’s a ‘one off’ or it never occurs again, there is an implicit cost / benefit assessment done (again, sorting data based on experience or judgement) that results in no formal measurement of the situation. We leave t up to the experience and the judgement of the manager to identify the tipping point where the scenario should be formalised.

    It is at that point that the data starts to be measured and can therefore be better managed to a more efficient outcome. This will allow the manager to move onto the next level of complexity. Lather, rinse, repeat. Measurement is always occurring, it’s just not always formal.

  3. I think there is this thought out there that you can only measure with quantitative data (numbers and stats such as from a survey or ticketing tool). You can also measure with qualitative data (interviews, focus groups, analysis of written material, etc.). It may be this data is measured differently but it is still being measured.

    I think the quote (misattributed or not) has a LOT of value, no matter which way you measure it. If you are not measuring it, it cannot be managed. How you measure it depends on the best way to get the data you are trying to manage and understand.

  4. I’ve always seen data as the compass point. It is not the entire map and even today not fully turn by turn directions. You need to mix the anecdotal with the data to get the bigger picture. Anecdotal information may give you the better sense of the terrain where the data compass point is indicating you need to go.

  5. I’m not sure what in the world you are talking about. Is it a semantic comment on the wording of the quote? Especially today, with “big data” and rapid data processing, looking for critical manufacturing variables in a “meta sense,” without the intervention of a human, as they trend over high volume production is an invaluable proactive tool for keeping a process in control. Only after the system identifies in real time non-random behavior trending toward the upper or lower control limits would a human be notified; in time to find and correct root cause before defects are produced.
    In my opinion, Dr. Deming would salivate at this measurement opportunity and capability.
    Respectively,
    Tom Borkes

    1. I suspect even with the wealth of data riches, Dr. Deming would recognize that not everything important can be managed. And that managers still have to make decisions in the absence of data… Data isn’t “meaning”…

      1. Managers always have data where they make their decisions. It’s axiomatic. You’re presented with a situation you haven’t seen before and you make decisions. Subconsciously you have identified the portions of the scenario that seem familiar; equally subconsciously you have filtered down your choices to a smaller solution set, from which you use ‘judgement’ or ‘experience’ to make the decision. You have the data and you’ve acted on it. It’s just not a formalised solution. But you’ve “managed it as best you can” and at some stage “we should circle back” to see if “there are some learnings to take away from the table”.

        I think what would annoy Dr Deming is the idea that the solution – however complicated – could not be inducted back into a formalised set of responses. In many ways this is EXACTLY what we want AI to be doing organisations. Filtering the ‘judgement’ and ‘experience’ into formalised solutions that can be programmed and replicated.

  6. I’m not sure what in the world you are talking about. Is it a semantic comment on the wording of the quote? Especially today, with “big data” and rapid data processing, looking for critical manufacturing variables in a “meta sense,” without the intervention of a human, as they trend over high volume production is an invaluable proactive tool for keeping a process in control. Only after the system identifies in real time non-random behavior trending toward the upper or lower control limits would a human be notified; in time to find and correct root cause before defects are produced.
    In my opinion, Dr. Deming would salivate at this measurement opportunity and capability.
    Respectively,
    Tom Borkes

Leave a Comment

Your email address will not be published.

Scroll to Top