Creating a Measurement Culture
By Dick Grimes
The phrase “a measurement culture” usually makes people think of places where tangible objects are produced that we can count, inspect, reject, or ship to buyers. We can measure (count) the number produced; measure (inspect) them according to some quality criteria (Are there blemishes? Does everything work properly?); and measure (count again) how many we ship to waiting buyers.
Or, they may think about measuring how late people are getting to work (the time clock); measure attendance (do we actually see them at work), or measure our seniority by checking off days on the calendar.
In reality, a measurement culture is much more than just the tangible items we see everyday. Its roots, from an organizational development perspective, are found in two very subtle, but critical questions:
How skilled are we at defining our work performance expectations?Do we, as an organization, know the difference between being “productive” and just being “busy?” The need to improve skills in defining work performance expectations is more than the typical vague examples we hear at work everyday such as:
“Make it look professional or ‘Act professionally”, or ‘Dress professionally” (What does “professional” look like? Does ‘professional’ have a specific color, shape, wording, flavor, scent, etc? How is the employee supposed to know what the speaker `expects as ‘professional?’)”You need to be more of a team player!” (We are always hearing about teamwork but what do we really mean? Does it mean I offer to help even if you haven’t asked for it? Do I work faster with my work so I can help those who work slower with theirs?)”Make sure they get their money’s worth.” (So, do we give them 5% more than they were expecting? Do we offer a guarantee or an extended guarantee? )”I will need that report in a timely manner.” (By when? In what format? Must it look ‘professional?’ Is on time more important than accurate?) The inability to define performance expectations well enough that someone can provide what is expected the first time can have a corrosive impact on morale. If the speaker cannot clarify their expectations so the listener has a decent chance of providing what they want, they have sown the seeds for disappointment, rework, and declining morale as the listener (often the employee) tries over and over, again, to hit a vague and seemingly arbitrary target. At some point, the employee thinks, “Why bother”, quits trying to get it right and just does the least they can to get by.
This article is not about those typical conversational requests that have been covered in countless training manuals and discussions about goal setting. Instead, it is about something more essential to the development of a confident and productive workforce: the ability to create meaningful measurements for critical workplace behaviors such as pride in work, team work, and customer service.
If we can devise a method of measuring those behaviors which result from an employee’s internal drive to excel, we can make sustainable progress toward developing our workforce.
When you take a moment to look at those topics, it becomes evident that they primarily reply on someone’s opinion although we will certainly include measurements where possible. For example, “pride in work” is more meaningful if you can understand the basis of their opinion with some measurement included.
Before we go farther, we should understand something about subjective and objective assessments.
Subjective is very open to someone’s opinion and may be difficult to defend or to reach a compromise when negotiating. You may have a favorite song but may not be able to convince someone else that it is great because he or she may not share your values and have a different opinion of what makes a song “great.”
However, if you said, “The song is by X, it is at #1 on the charts, and they have had 13 million sellers in the past five years”, you are making a strong argument with some objective points included (measurable) that it is probably worth listening to by the other person.
When we devise a measurement scale for subjective topics, it is important that we include as many elements as possible on which people can agree – especially the people most affected by your measurements.
It is easy to measure if you will work with your employees to define jointly the behavioral traits that demonstrate the topic. We specify behavioral traits because people are more likely to agree on what they saw (objective) than on what it means (subjective). (Once again, it is critical to the success of this measurement process that you include those whose behavior you expect to measure. Their “buy in” is critical for the success of this approach.)
For example, if you wanted to measure “PRIDE IN WORK,” you first identify behavioral traits (things they do that everyone can see) that indicate they have pride in their work. Do not ask for things that are invisible such as “Pride in work means I feel good about myself.”
Some behavioral traits they suggest may be like these:
There are no spelling errors. (This is objective because the words are spelled correctly or not. There is little room for argument.) It is always on time or before. (This is objective because it is on time or not. There is little room for argument.)The person always uses the proper format for the report. (This assumes there is an existing format available to follow for the report. They either use it as is or they do not.)There are no smudges or typos on the form. (Either errors are there or they are not.) The data is always accurate. (It is accurate or it is not. There is no middle “subjective” ground.) Note: Do not spend time worrying if these are the “right” traits to consider because, if everyone who is involved in the assessment agrees these are the “right” topics and there are no legal, procedural or ethical violations, then no one is harmed with the outcome. Therefore, if we all like them and no one is harmed, THEY MUST BE THE RIGHT ONES.
Establish a scale with a range of values from one extreme to the other for use with each trait. There is no magic number of survey points that you should use (1-3, 1-5, 1-7, 1-9 ? But keep in mind that attitudes are not as precise as tangibles and too many may be meaningless or confusing. (This scale of attitudes is a Likert Scale.) For example, something like this is just asking for frustration from those taking the survey: 0 = never, 1 = rarely, 2 = more than rarely but not often, 3 = often but not frequently, 4 = frequently but not daily, 5 = daily
Make it easy for people to use your survey because if you do not, you will not have any results to collect!
Also, include a way to respond if someone has an answer that does not fit the criteria of the question. For example, look at the survey (below) that asks for the frequency that an observer sees a behavior.
If the observer does not think he or she has seen enough to make a fair observation, give them a chanced to say, “Does not apply” or something like that, so they do not feel compelled to force an answer and select a number that does not truly reflect their situation. If you mix their forced answer with others that are valid, you will only diminish the reliability of your other numerical scores.
1 = I never see this (Do not get overly concerned with the definition of “Never”. As long as you and the employees agree on what frequency it means, the scale will work.)
2 = I occasionally see this (Do not get overly concerned with the definition of “occasionally”. As long as you and the employees agree on what frequency it means, the scale will work.)
3 = I always see this (Do not get overly concerned with the definition of “Always”. As long as you and the employees agree on what frequency it means, the scale will work.)
X = Does not apply
CAUTION: Professionals such as engineers, surveyors, and accountants are used to working with specific data and may be uncomfortable with this concept. They should remember we are measuring impressions and attitudes that may be vague at best. Asking a customer to rank satisfaction along a scale from “Not satisfied,” “Some satisfaction”, Very Satisfied,” or “Does not apply” is more useful than asking them “are you 43% satisfied or 72%?”
Then create a scale that converts the average of the trait scale (#2 above) to your performance assessment system. Joan Smith’s Pride in Work (“Pride in Work” is the behavior we are evaluating) Evaluation
There are no spelling errors. =3
It is always on time or before. =2*
She always uses the proper format for the report. =2*
There are no smudges or typos on the form.=3
Her data is always accurate. =2*
Average score for the observation period =12/5= 2.4
*A wise leader will have documentation of the times when the employee did anything that would result in a less than perfect score. You can expect them to ask for proof they did not earn the top score.
In addition, you will improve their morale if you put the burden of proving they did not get the max score on you instead of putting the burden of proof on them they did earn it. This would be like a teacher in school telling you at the start of the year “Everyone has an “A” in this class until your scores show otherwise.”
The result from a grade standpoint is probably the same but the attitude of the students is much more optimistic and they may score better!
These five “behavioral traits” from part #1 define the behavior “Pride in Work” and the rating scale for each trait (part #2) gives us a way to measure it. Then we can average the results from the survey given to people who see’s Joan’s work and put them into our performance assessment grid in Part #3 for the behavior “Pride in Work.” We can compare her overall score of 2.4 to the performance rating scale our organization uses and see that a 2.4 means, for this trait, she has a “satisfactory performance with lowest merit increase.”
Example of a performance scale range using a 1.0 – 3.0 scale:
1.0 – 2.5 – 2.8 = above satisfactory performance with medium merit increase
>2.8 – 3.0 = outstanding performance and maximum merit increase
If she wants to improve her score for the next evaluation period, she can look to the rating grid (#3) where she got scores of ‘2’ instead of ‘3’ and see that she needs to work on being on time, the report’s format, and accuracy.
“Is allowing employees to rate each other in teamwork a good idea?” Ask that question among a group of supervisors and managers and you will get many reasons for and against it. We believe that it is a good idea only when the group doing the rating has the maturity to understand the benefits it offers.
Here are some issues for consideration about the argument for those who have not considered it before.
Teamwork is important to the success of our group. (No one will argue that fact.) There are only two ways to evaluate individual performance contributions by members of the team: the leader can do the evaluation or the fellow team members can. The leader is not always present while the team is working. Frequently there are duties unique to leadership that requires him or her to be elsewhere such as attending meetings, working in the office planning, scheduling, budgeting, reporting, etc. There may be team members who work more diligently while the leader is present than they do when the leader is absent. Other team members frequently have to pick up that slack to meet the group’s production goals. Since the leader was periodically absent and not able to observe performance at all times, there is a distinct possibility the slacker will get a performance rating that is higher than deserved when the leader does individual performance assessments on group members. This is unfair to those who had to pick up the slack to meet the group’s goals. The group’s members can solve this problem by doing anonymous assessments of each other if they choose to act objectively. The leader can use these for the individual’s performance scores under the “teamwork” category. Some group members will object saying that assessments are the job of the leader. While that is true, it is also true that the frequently absent leader cannot evaluate individual performance within the group as often as other members can. The group can choose to score itself anonymously and be very accurate or insist the leader do the scoring and admit there may be undeserved scores. They cannot have it both ways unless the leader gives up some leadership duties. However, if the leader stopped doing those leadership duties, he or she would not remain a leader very long. The possibility of groups of team members giving those they do not like lower scores (or friends higher scores) than they deserve can be defeated by using the Olympic judge technique of throwing out the highest and lowest scores. Alternatively, you can require specific examples be included for the highest or lowest scores to count. If a group is willing to self-score, work production will increase because the slackers can no longer “get away” with their games: they know it and the group members know it. Finally, this is very similar to a 360° assessment frequently used by human resource departments to assess mid-to senior level managers. The concept is NOT that unusual.
So, the first of the two critical questions was: if the organization is not used to defining expectations in clear and concise ways, how can they expect to ever develop their workforce into something more than it is now?
We have just demonstrated an easy way to use specific behaviors as the definitions for the seemingly subjective topics of pride in work, team work, and customer service.
Which leads to the second point: if they cannot define expectations sufficiently to distinguish between “productivity” and just being “busy”, they will always have a very inefficient, expensive, and non-productive workforce.
But, if the culture realizes the basic criteria for measurements are quality, quantity, and time, it becomes much easier to define expectations and get what they want more quickly without rework. Quality, quantity, and time (QQT) tells us how good it must be (quality), how many there must be, and when we need them.
This, then, is the difference between being productive and just busy. Productivity occurs when we know how many, how good, and by when. If we don’t have those three elements, we are just killing time with busywork.
For example, a clerk that spends three hours (time) filing a stack of documents (quantity) isn’t very productive if there are many errors that must be corrected (quality). Clearly, the clerk was busy but not very productive.
Another example may be wasting time arguing with an employee about what constitutes a “professional appearance.” A measurement culture does not reply on, “I’ll know it when I see it” but, instead, defines acceptable attire with measurable terms such as long sleeves or collars (visually measurable) or collars (visually identifiable from a sample drawing.)
Let’s try a pop quiz: Read each situation and decide if the person is busy or productive. Remember, we define “productive” as having all three elements of how much (quantity), how well (quality), and by when (time) present in a situation.
If “busy,” what is missing? Grant has been pushing the cart loaded with statements up and down the hall for the past 45 minutes.
Logan has processed 200 documents in the past 3 hours.
Murphy has the machine running items at a speed of 375/hour and a reject rate of .82%
Cindy completed 3 hours of classes at night school this past quarter.
Grant is just busy. Nothing has been accomplished. What is the “quality” measurement? Logan has processed 200 documents but are they the correct ones? Where they processed accurately? There is no QUALITY component Murphy has been productive. The quality component is the “reject rate”. Cindy completed the courses but we do not know her grades (quality). If she failed, there is no productivity. If she received an “A”, she was more productive than if she received a “C”. The quality, quantity, and time concept is relatively simple and we do not need to elaborate more here. When an organization is able to incorporate all three elements into its expectations every time almost without thought, it is well on the way to creating a measurement culture which is a foundation for an effective workforce development initiative. The quality, quantity, and time concept is relatively simple and we do not need to elaborate more here. When an organization is able to incorporate all three elements into its expectations every time almost without thought, it is well on the way to creating a measurement culture which is a foundation for an effective workforce development initiative.
Richard (“Dick”) Grimes uses his 30+ years experience in training and operations management for private and public organizations as a foundation for his company, Outsource Training.biz LLC (http://www.outsourcetraining.biz).
Through Outsource, he creates and delivers workforce (project team) development courses and consults in areas of employee performance, leadership, and organizational evolution. The quality of his courses has earned Outsource “approved education provider” status by two major and distinct professional certifying organizations: National Council of Examiners for Engineers and Surveyors (registered provider #N0006) and the Human Resources Certification Institute (provider #2299).