The First Commandment of Performance Appraisal

By on July 28, 2012 in Current Events with 32 Comments

In my most recent article, I addressed the subject of critical elements.  These are the rating categories used in performance evaluation systems.  I pointed out how generic elements and elements that presume to rate competencies inevitably lead to subjective standards and ratings.  Such evaluations do not serve to improve performance and, therefore, are of no particular benefit to the agencies using them.  Moreover, rating these elements often leads to allegations of bias and favoritism.

In this article I want to discuss performance standards – the measures used to evaluate each critical element.  In doing so, readers can discover what I believe is a fundamental law of performance evaluations and the most common reason appraisals have failed us over decades.

Law, regulations, and generics

Standards are often referred to as “measures” or “yardsticks”.  They inform employees of the expectations to be met in a given element.  Our Office of Personnel Management (OPM) oversees the government’s performance appraisal system.  In the Code of Federal Regulations (5 CFR 430.203), they define performance standards as follows:

Performance standard means the management-approved expression of the performance threshold(s), requirement(s), or expectation(s) that must be met to be appraised at a particular level of performance. A performance standard may include, but is not limited to, quality, quantity, timeliness, and manner of performance.

The idea was to have individually written performance standards (or expectations) for each critical element.  This challenges managers to come up with specific measures for specific jobs under their leadership.  In fact, the law itself (5 USC 4302) reads, in part:

Under regulations which the Office of Personnel Management shall prescribe, each performance appraisal system shall provide for establishing performance standards which will, to the maximum extent feasible, permit the accurate evaluation of job performance on the basis of objective criteria (which may include the extent of courtesy demonstrated to the public) related to the job in question for each employee or position under the system…

This isn’t happening in most Federal agencies.  Many HR departments have devised “generic” or “benchmark” standards for rating purposes.  Their intent is to relieve supervisors of the chore of defining successful performance.  They also believe that providing everyone with the same rating elements and standards gives an appearance of fairness and equity… despite the legal mandate for objectivity.

Benchmarks and guesswork

Pre-written performance standards are intentionally vague so that they can be used by any supervisor for any critical element and any job government job – from Fork Lift Operators to Research Scientists.  Such “canned” standards inevitably lead to subjective ratings, which might have been the same had there been nothing at all in writing.

Most of the supervisors and managers I meet (from State, Navy, Agriculture, Interior, Army, etc.) are trying to rate their workforce fairly and accurately.  Despite good intentions, however, their ratings are based only on anecdotal evidence… and little of that in most cases.  The generic benchmarks encourage subjective evaluations.  Some agencies attempt to describe high, middle, and low levels of achievement; however, the language they use must be interpreted and applied by supervisors who are often baffled by it.

Standards as they were meant to be

In seminars, I ask managers, who are required to develop their own standards or supplement the generics with their own criteria, to consider the first three techniques offered up by OPM in the Code of Federal Regulations – quality, quantity, and timeliness.  These are the traditional outcome-based measures of performance that senior managers, consultants, and HR specialists continue to advocate.

I ask seminar participants (and now you) to think of all the ways a real supervisor would actually know if an employee is delivering quality, quantity and/or timeliness.  Examples of performance measurement techniques they suggest are:  supervisory inspection/observation; customer feedback; coworker feedback; review of logs; and returns/rework.  Of course there are more.

Take another moment to consider ways a supervisor could know if you or another employee is delivering results – quality, quantity and/or timeliness.  This is what management is supposed to do to have objective appraisals.

Note to HR specialists:  Something becomes noticeable during the course of this exercise – perhaps for the first time.  Quantity of work performed may prove utterly unreliable as a measure.  As managers try to come up with real-world measures for quantity, they commonly give up.

Quantity usually depends on external factors such as what other jobs are being assigned to the employee, how many requests for work come through the door or across an employee’s desk.  Moreover, different assignments have different levels of complexity.  This makes quantity measures particularly unreliable.

There is also the obvious connection between the quantity of work performed and the time it takes to perform it… and the more quantity and timeliness are stressed, the worse quality is likely to be.  As the old saying goes, “The faster I work, the behinder I get.”  Measuring outcomes has many implications that need to be considered by those who develop performance standards.

…and now, the Commandment

Beyond the obvious problems with performance measurement, there is this one:  every technique or idea you might imagine for assessing quality, quantity, and/or timeliness leads to a significant workload for the front-line supervisor.  When evaluating results using any of these measures, s/he must regularly observe and log individual performance throughout the year.  What’s more, records must be kept on each individual supervised… in each critical element.

This understanding leads to the First Commandment of Performance Appraisal:  If thou attemptest to rate employees in terms of Quality, Quantity, and/or Timeliness thou shalt use metrics that require thee to observe, log, and keep book for every employee – in every critical element.

Supervisors who lack elaborate reporting software (such as that used in call centers) don’t like reading this.  The Commandment, however, doesn’t end there.  There’s a corollary which explains why agencies use benchmark and generic standards: If thou cannot or will not keep a bean count on individual employee performance in several critical elements, thou shalt rate thy employees subjectively.

Where metrics aren’t readily available, then a “beauty contest” is the inevitable result.  Some supervisors keep evaluation notes throughout the year.  Most don’t… nor do they have any memory of when the last rating year ended and this one began.  Ratings become guesswork.

Get real!  Get in GEAR!

Managers tell me that there aren’t enough hours left in a day to observe, log and keep individual records on each employee.  The say that time for evaluating employee performance comes after the meetings, “special projects”, and reports that government demands of them.  Moreover, they find “bean counting” (of deadlines met/missed, errors, etc.) demeaning and/or distasteful.  Most managers I meet question whether their agencies would actually benefit from such an investment of time and energy.

The National Council on Federal Labor-Management Relations is grappling with these issues with its GEAR model – now being piloted in several Federal agencies.  GEAR (Goals, Engagement, Accountability, and Results) is initially focusing on the infrastructure needed to rate workers in results.  This includes: regular supervisory feedback; holding supervisors accountable for making appraisals a priority; serious training in performance management; and reconsidering the selection process for new supervisors/managers.  I commend them for putting the horse in front of the cart, and wish them luck with such an ambitious undertaking.

It’s a Commandment – ‘fess up to it!

By my reckoning, if evaluations are to prove useful, the Commandment needs to be acknowledged by OPM, senior management, the National Council, and Chief Human Capital Officers who are responsible for making the law and regulations work.  For decades, front-line supervisors have on the receiving end of rhetoric regarding “results-oriented” and “objective” measures without sensing a commitment from those at the top to do so themselves.

Those who advocate for objective and results-oriented standards need to explain how and why supervisors and managers should adhere to the Commandment.  Those who have tried to quantify the work of Economists, Electricians, Biologists, and Law Enforcement Officers have been frustrated for years.  In some cases it seems as if senior management and HR are more focused on finding something to measure than on actual job performance.

Here’s the Commandment again: If thou attemptest to rate employees in terms of Quality, Quantity, and/or Timeliness thou shalt use metrics that require thee to observe, log, and keep book for every employee – in every critical element.  Now consider a sign that hung in Albert Einstein’s office in Princeton: “Not everything that counts can be counted, and not everything that can be counted counts.”  I think there’s wisdom there.

Subjectivity reigns

In light of the commandment, it might be better to simply acknowledge that employee performance ratings will commonly be subjective.  Where metrics are available and work to motivate, go ahead and use them.  Insisting on specific, objective, measurable, outcome-driven standards where such data is unlikely to be harvested puts too many supervisors in the awkward position of fudging.

The bond of trust between a supervisor and subordinate at the workplace (where submarines are repaired, veterans treated, forest fires suppressed, contracts examined, roads designed, and nuclear materials safeguarded) is paramount.  It is jeopardized when management fails to practice what is preached.  Insisting that metrics reign and performance ratings are science rather than art is one area where rhetoric and reality don’t match up.

Gathering data, analyzing results, and aiming for continuous improvement is a worthy endeavor.  As a management practice it can help all of us see where we are and where we want to go.  Metrics relating to quality, quantity, and timeliness are at the heart of most management philosophies like MBO, SQC, FTF, and TQM.  But the First Commandment of Performance Appraisals hasn’t been followed by managers in most agencies.  That’s why we see so many canned standards.

Offering up “benchmarks” and “generics”, while insisting on results-driven performance standards, isn’t fooling anyone.  It will take honest adults to recognize and acknowledge the contradiction. Perhaps the National Council on Federal Labor-Management Relations can help in this regard. A little candor might go a long way.

© 2016 Robbie Kunreuther. All rights reserved. This article may not be reproduced without express written consent from Robbie Kunreuther.


About the Author

Robbie Kunreuther is the Director of Government Personnel Services (GPS). GPS provides 1 to 3-day seminars to Federal agencies in four subject areas: Dealing with performance and conduct issues; Developing sensible performance appraisal criteria; Fostering cooperative labor-management relations; and Applying mediation skills in the workplace. Over the years, Robbie has trained thousands of Federal supervisors, managers, HR specialists, and union officials. For more information about him and GPS, go to

Post a Reply

Your email address will not be published. Required fields are marked *

32 Replies

Comments RSS

  1. Lateen85367 says:

    …but it should not be an end of year surprise.  

  2. Jonah B. says:

    “If thou attemptest to rate employees in terms of Quality, Quantity, and/or Timeliness thou shalt use metrics that require thee to observe, log, and keep book for every employee – in every critical element.”

    Oh, what nonsense. I used to manage senior network engineers… how do you assign sensible metrics for work that varies constantly and is not on any remote production schedule? Work that completely defies metrics, in other words?

    For a time our management became metric-happy and we were forced to come up with utterly silly, extremely arbitrary and unenforcable metrics. In fact, our aganecy even hired a contractor to help us with this! All for naught.

    Your First Commandment is a false religion.

  3. Short-Time Prawn says:

    The crusade for a Perfect Appraisal System is like “…trying to catch the Devil’s herd – scross the endless sky.”

  4. Jarvis32 says:

    Every year, in DOL, the elements are used as a huge dumping-ground for the increasing extra work and responsibilities forced on employees as staff retire or resign. All the extra work is considered ‘part’ of one element or the other for many. Getting ‘meets’ because of it, looks like staff are just meeting the minimum of what is expected, not that the workload doubles every year. It is all according to the supervisor. Some staff members receive rewards for what they do, others do not. Very biased procedure.

  5. ER/LR Specialist says:

    I’ve worked with performance management in the federal government for many years and fully agree with the positions Robby has taken.  One area he missed though is the one Underpaid CS addresses.  Responsibility should also be on the employee to identify what accomplishments he/she has achieved during the year in the achievement of identified organizational goals.  

  6. Joejacs says:

    PLEASE send this article to the Congresspersons who insist on “Pay for performance”.  I  have dealt with the problem of measurable critcal elements, and the age old question of what is the highest priority?

  7. Retired LR says:

    As a former LR ER Manager…let me say clearly that performance management in the federal sector is a JOKE. I have seen many many times….managers who boost the nonperformers because they are part of the team or should I say clique…those that are a target of jealousy, threatening professionally etc seem to fare worse even though those employees were diligent , preserving and professional…..I have seen unions raise absurd arguments to defend the losers also…so here you have it..both managers and employees can be losers…results oriented performance management can be manipulated……now…the answer…here it is….do away with the performance management the way we know it in the federal sector…replace it with this….employees need to convince management by clear documentation each year that they should be retained because they saved money for taxpayers and were ETHICAL…let me repeat ETHICAL.  Now one would argue that some jobs have no savings standards…bull…even Border Patrol, TSA agents, VA cafeteria workers can articulate some way of overseeing fraud waste and abuse..novel idea huh? Aawards…well they can receive 5% of the savings…motivation enough?

    • Retired Manager says:

      The summary of these comments is that the federal performance management system does not work.  So why waste time trying to make it work?  Without spending a lot of time, you can put together a reasonable (albeit subjective) rating system that has the required buzz words and metrics, so that the HR folks, the lawyers, EEO and upper management are satisfied — and although the ratings distribution will not be a bell-shaped curve, you will have met the requirement.

      So, how then to correct poor performance?  Easy.  Every poor performer eventually messes up on conduct, and repeatedly so.  Forget performance-based action.  Go for conduct-based action every time.  As a supervisor, I successfully removed 10-, 20-, and even 30-year tenured poor performing employees by going the conduct route.  It takes a fair amount of documentation, but it is all objective, as opposed to the subjective performance-based actions.

      My favorite charge?  Failure To Follow Written Instructions.  They’re messing up?  Once you determine it is not a training issue (“Can” they do the job?), but is instead a conduct issue (“Will” they do the job?), then you give ’em an instruction letter.  And you’re off to the races.  After the third or fourth cycle of “written instructions”, “failure to follow” counseling session, progressive (increasingly stronger) penalty — you take removal action if the person didn’t get the message.

      If you’ve partnered with your HR, Legal and EEO staff, your documentation is in place, and you have a “Deciding Official” who will support you — you can issue the same “wake-up call” to a poor performer as you would hope to do with a Performance-Based action, up to and including Removal.

  8. steve5656546346 says:

    Measuring performance empirically (without subjectivity) is impossible.  

    How did you set the standard?  By guestimate, since your budget didn’t allow for a million dollar study.But now the standard is set (with what ever degree of accuracy), and the rating year actually begins:  and things change.  The work within that standard may get more difficult or easier.   A new supervisor may prove to be more competent in spotting errors, or less competent. There could be other changes as well. But work outside of that standard affects that standard too:  changes may mean that there is less time to do the work in that standard…or more time.  New difficulties may add to the level of stress and exhaustion, or reduct it.Well, you an CHANGE the standards:  but how?  You again reach the difficulty of coming up with the right measurement and the additional problems of continuing change.

    OK, so why not just accept that there is some subjectivity?  Because of EEO complaints (without regard to whether or not they are justified).

    There is no solution:  which is one of the countless reasons that the government doesn’t work because it can’t work.

  9. Klw1953 says:

    After 30 years of government service as well as years of private sector work experiance I never understood all this government rating stuff.  In ALL my private sector work, you either perform the job you were hired to do or you don’t, period.  I also never understood why it takes so many lines of government supervison actually performing none of the real work but strictly supervisors and supervisors of supervisors called managers, deputies, front line supervisors and all the others in between the employees actually performing the mandated work to the Secretary of whatever department it is.  Seems to me like if so many supervisors are needed someone isn’t doing their job!

  10. Lathe Ragels says:

    Nicely done as always Robbie!  You pointed out basically all of the reasons why most government employee appraisal systems fail. 

  11. Troy1967 says:

    If you can’t write objectives and measures for average and good performing employees, how can you possibly write a performance plan for a poor performer.  And even if you do write a performance plan for the poor performer, how do you defend yourself as a superisor against the allegation that you are playing favorites among your employees, if the other employees don’t have defensible plans?  One way to approach the problem is to focus on attributes as well as deliverables.  For example, in writing objectives for a GS-12 use attributes for a GS-13 for the “above satisifactory” standard to describe the level of detail in task description,  independence of task perofmance, and degree of supervision required, AND attributes for a GS-11 for the “below satisfactory” standard.  As a supervisor, you are not only interested in getting the results of the completed task, but getting the result with the appropriate level of effort on your own part.  If you have to continually push one employee to get the desired result and another employee goes off and does the task on his own, you can’t explain a difference if rating if the rating is only based on the deliverable.

    • steve5656546346 says:

      The performance appraisal process is a myth, but MSPB buys into the myth–which, by law and regulation, they must do.  So, you put somebody on a PIP based upon the standards which can be no more than a guess absent a million dollar study…and perfect knowledge of the future…  And the MSPB buys it.

  12. Seadog says:

    Generics never work, especially bean counting, unless every job is exactly the same.  Poor employees in management greatly amplify the ratings problems when the rating criteria isn’t flexible enough to be applicable; they also often don’t care to do the work needed to make effective ratings happen since it doesn’t affect their own ratings.  Until there is an effective system of retribution for poor ratings, and also for those crying ‘wolf’, there will never ratings truly reflecting the quality of work and employees.

  13. USDA/FSA manager says:

    Using USDA created standards and documenting substantial performance problems with one employee during the past 2 mid-year and summary reviews resulted in only a letter of counseling that took over 6 months to be approved by my agency’s HR performance specialist in Kansas City. Agency used terminlogy of 1-3 mistakes per 100 contracts was useless for anything disciplinary as the HR specialist said it was too vague to be used. Thank God that the person now is having conduct issues that can be delt with properly. Performance plans are not worth the paper used to print them.

  14. Stellamaris says:

    Every year I hear from employees about their evaluations.  When they tell me about a low score, I suggest they ask their supervisor for specifics.  Inevitably, the supervisor fails to come up with specifics.  Years ago, as a teacher supervisor, I consistently advised the teachers to create anecdotal records.  It baffles me why supervisors fail to do the same.  Then when performance review comes, they find they created a difficult situation because they can’t back up what they wrote.  Unfortunately, this recently happened to me.  In a move to get create a constructive termination, they reassigned me to duties in a field I had no experience.  I was forced to become an HR Information Technology Specialist from a role as a Human Resource Development Practitioner, a role I worked in for over 40 years.  So at 65 years of age, I was supposed to learn programming and IT duties.  When time for review came, I was told that I needed to come up to speed.  However, management failed to point out specifics because there reall was no data to support the review. 

    • Old Fed says:

      That is blatant age discrimination, as I am sure you are aware. File an EEO complaint. But at age 65, with that many years of service, I would just retire if I were you. Why put yourself through the aggravation? Just not worth it.

  15. Jonah says:

    My government entity – in an effort to “partner” with the Union – has created all sorts of programs (flexitime, flexiplace, telework, individual flex time, etc.) which make it virtually impossible for me to know when my employees are even on the job, let alone know how they’re doing. Are they working 40 hours a week, really? I have no idea. 

    • Mad Hatter says:

      You don’t know how they’re doing?  Aren’t they responsible for producing work that you can measure them by?  Or do you just measure them by time in the office?  I’m no great fan of telework, but you don’t seem to understand supervision.

      • Bill T. says:

        Really, that’s where the rubber meets he road, isn’t it?  If someone is in the office 40 hours and never produces a deliverable or another one puts 30 hours in and meets their goals, who’s the under-achiever? Tell me what I need to get done, stand back and let me work.

    • NormaRae says:

      And nor would just because someone is sitting in an office all day – surfing the net, forwarding useless email “chain letters” that clog inboxes, daydreaming, or – worse yet – bothering other people who ARE trying to work by “stopping by to chat.”  It is inane that managerment believes that mere physical presence = ability to control productivity.  Unless your job demands physical presence (like working a forklift, for example) there is really only one way to know if someone is working 40 hours a week.  It’s called “output.”

  16. Bill T. says:

    What about the peoeple who are very good at what they do but are very poor at self-evaluation. The English majors may poo-poo that and come up with all kinds of reasons why that shouldn’t be the case but they couldn’t even begin to perform the tasks I’m getting paid to do.

    Rated down for poor self-evaluations vs poor actual performance.

  17. msgrowan says:

    I remember an occasion back in the mid-’80s approximately, when the justly celebrated Dr. W. Edwards Deming was asked to address a group of OPM senior managers on the issue of the performance appraisal process in the Federal government and what OPM could do to help agencies to improve their systems.  A videotape was made of the session, and I remember seeing it subsequently.  He started off his comments by making the bald statement” “You have an impossible job!”  He went on to say that the goal of an objective, equitable, effective, and efficient performance appraisal process was impossible to attain, and that the attempts to impose one were inevitably divisive and toxic to workplace cohesion.   With all the barrels of ink – liquid and electronic -that have flowed in the interim on this topic, and all the no doubt millions of manhours devoted to this end in the intervening quarter century plus, in my opinion nothing has resulted that would call into question Dr. Deming’s key insight on this issue.  Unfortunately, Federal law requires such a circle squaring program, and so we will continue to go on with such counterproductive program, which only perpetuates hypocrisy and attempts to portray ongoing failure as actually resulting in an effective performance appraisal mechanism. 

    • HRguru says:

      I guess.  But don’t companies deal with these same issues?  Why is this a particular problem for the Federal government and not for other entities?  It seems like all an excuse for managers to throw up their hands.  

      • same old, same old says:

        But they dnt just throw up their hands.  They use the performance management process to reward the favored few.  I’ve shown that my work was more accurate (I can actually use grammar correctly), but then the manager decides the few cases the other person worked were somehow more complex.  I’ve read their memos, and there was nothing complex about them.  Maybe the government cannot figure out how to effectively rate employees because that keeps OPM and HR fully employed developing and rolling out new systems (or old ones under new names) that are destined to fail. Full employment for HR:  good; fair and transparent employee evaluations:  bad.

      • steve5656546346 says:

        The difference is that the private sector is not nearly as litigious, and they often don’t have all of the straight-jacket regulations.

        Subjectivity is a proper and necessary part of the performance appraisal process.

  18. Underpaid CS says:

    really nice write up but…

       I don’t need a supervisor looking over me and measuring work that he knows nothing about or couldn’t do it due to his lack of experience… so I log my own work and store it where everyone can see exactly what I did – and write up my own evaluations – my supervisors job is to put on a tie everyday at work…and rate my work as exemplary – and since he trusts me it works perfectly and makes his job easy compared to mine … I’m a doer of actual work… he is a figurehead… and writes up the standards and then I exceed them… so call me an inferior subordinate… very insulting since I am the one doing the ACTUAL work…

    • Mad Hatter says:

      Your post reads like a Dilbert strip. I’ve known many people who had your view of supervision until they became a supervisor and then said it was the hardest thing they ever did.

    • Edsclone says:

      Of course, if you are the only worker doing the job it works perfectly.  The problem comes when there are several of you doing the same thing.  How can a manager/Supervisor who many times than not is not experienced with the job at hand, be forced to evaluate any of his workers?  The responsibiliy of evaluating an employee should fall on the immediate Supervisor or Lead Man who must be experienced with all jobs assigned to the group and anything else related to the particular operation.  However, as in anything in life when dealing with humans there is no way to take away subjectivity, as sad as it might be.   

    • EWeaver says:

      Former Baltimore Orioles manager Earl Weaver was once asked by a reporter during a post-game interview whether he was having fun that season.  Earl replied:  “Having fun?  Ask my players – THEY have fun.  I’m the manager – that’s work”.  With more than 10 years of technical work as an engineer, and even more since then as a manager, I couldn’t agree more. 

      • Underpaid CS says:

        Earl Weaver got paid so much –several times over what the POTUS makes – I would say that would be “fun” enough and baseball is just a GAME.  So is the performance appraisel process.  The smart ones know how to play it best and get the highest ratings where I work.  I’m going to say it again MY BOSS doesn’t have the competency to do my JOB.  That goes for the division chief also.