How To Measure Anything by Douglas Hubbard is a popular book at VMware Tanzu Labs. In it, Hubbard takes a strong, opinionated stance on both the concept of measurement, the value of measurement, and how to measure things that are considered immeasurable.
One of our team members published an old-fashioned book report on How to Measure Anything, which you can read here. Below we summarize one fascinating and helpful section of the book, along with our notes and thoughts, as this section is particularly helpful for the Product Valuation workshop.
First we want to highlight Hubbard’s analysis of why people often think things are “immeasurable.”
There are just three reasons why people think that something can’t be measured; these reasons are based on misconceptions about different aspects of measurement. We’ll refer to these misconceptions them as the concept of, object of, and method of measurement:
We’ll expand upon each of those misconceptions.
Implicit or explicit in all of these misconceptions is that measurement is a certainty—an exact quantity with no room for error. This is a fallacy. Some amount of error is unavoidable, and measuring can still be an improvement on prior knowledge even if the measurements have error. This concept is central to how experiments, surveys, and other scientific measurements are performed.
Unfortunately, allowing for unavoidable errors or room for improvement can feel contrary to some definitions of “measurement”. For example, Hubbard refers to this definition:
“Measurement: a quantitatively expressed reduction of uncertainty based on one or more observations.”
Note that this definition for measurement states a measurement is quantitatively expressed. The uncertainty has to be quantified, but the subject of observation might not be a quantity itself; for example, it could be entirely qualitative, such as membership in a list.
We must avoid the trap of thinking that all uncertainty can be a measurable quantity.
Regardless of which concept of measurement is adopted, some things seem immeasurable because we haven’t unambiguously defined the object of measurement.
For example, we might want to measure an “intangible” thing, such as a company’s public reputation. A reputation itself might be hard to quantify, but we can quantify the impacts of a company’s reputation, such as customer referrals or positive vs. negative social media references. Now we have begun to identify how to measure it.
The clarification chain is helpful in identifying how to measure the intangible:
In addition, it helps to state why we want to measure something; identifying the “why” can lead us towards the “what” is really being measured.
Here we will quote Hubbard:
“When thinking about measurement methods, someone may imagine a fairly direct case of measurement. If you measure the length of a table…you have direct access to the entire object of measurement. If this is the limit of what one understands about measurement methods then, no doubt, many things will seem immeasurable."
In other words, if we limit ourselves to only certain means of measurement, then anything that cannot be measured by those means might seem immeasurable. Avoid this trap by searching for other methods of measuring. A table, for example, has not only dimensions, but also weight, color, design style, aesthetic appeal, monetary value, or even sentimentality.
We hope this was a helpful summary regarding measuring the immeasurable. By expanding your concepts of measurement, re-thinking the object under measurement, and your means of measurement, nearly anything can be measured. Be sure to check out the Product Valuation workshop for practical applications of these concepts that can help your product development teams.