Chapter 4

Measuring Performance for Pay: 5 Different Approaches

The rat­ings approach

The clas­sic approach to mea­sur­ing per­for­mance for pay pur­pos­es has been for man­agers to allo­cate a rat­ing to each of their team mem­bers which then par­tial­ly deter­mines their pay or bonus out­come. There is no rea­son why you can’t still col­late per­for­mance rat­ings at the end of the year under a con­tin­u­ous per­for­mance man­age­ment approach, ensur­ing you make it a sep­a­rate process, as explained above.

How­ev­er, research stud­ies have shown that there is lit­tle cor­re­la­tion between per­for­mance rat­ings and actu­al per­for­mance. This is due to a com­bi­na­tion of man­ag­er bias and man­agers reverse-engi­neer­ing per­for­mance rat­ings to get the reward out­come that they want for their team mem­bers. This is one of the main rea­sons why sur­veys con­sis­tent­ly show that most employ­ees find annu­al appraisals to be unfair, despite the numer­ous hours of work that typ­i­cal­ly go into rat­ing cal­i­bra­tion. Research has also found that link­ing per­for­mance rat­ings to pay ends up demo­ti­vat­ing the major­i­ty of employ­ees, which iron­i­cal­ly is the exact oppo­site of what per­for­mance relat­ed pay is attempt­ing to achieve.

This has led many organ­i­sa­tions to adopt alter­na­tive approach­es to rat­ings and stud­ies are now find­ing these alter­na­tives to be more effec­tive than rat­ings. For exam­ple, a study of 22 com­pa­nies that had removed for­mal rat­ings and were between two and five years into their new per­for­mance man­age­ment frame­work found that employ­ee engage­ment went up in all of them after the new sys­tem was in place, and it has con­tin­ued to do so since.

Let’s explore some of the prac­ti­cal alter­na­tives to per­for­mance rat­ings when it comes to mak­ing deci­sions about pay.

The man­ag­er dis­cre­tion approach

When Gen­er­al Elec­tric, one of the largest organ­i­sa­tions in the world, adopt­ed con­tin­u­ous per­for­mance man­age­ment, they also decid­ed to get rid of rat­ings. In 2016 I saw one of their HR Exec­u­tives, Anas­ta­sia Kuzmi­na, present at the CIPD Per­for­mance Man­age­ment Con­fer­ence and she gave a com­pelling argu­ment for their new approach. She said:

Our man­agers were telling us that they are able to make mul­ti-mil­lion dol­lar deci­sions about busi­ness projects, yet they are not allowed to decide whether a team mem­ber should get a 2% or 3% pay rise. They felt it was a crazy situation”.

So now GE run an annu­al pay review process and give each man­ag­er a pay pot and the man­agers can deter­mine how best to spend that with­in their team. This allows man­agers to take a num­ber of fac­tors into account when decid­ing on pay – per­for­mance, poten­tial, risk and impact of leav­ing, changes in mar­ket rate, etc.

To min­imise sub­jec­tiv­i­ty, man­agers are asked a series of per­for­mance-relat­ed ques­tions to help frame their think­ing, such as:

  • How has this per­son made an impact in areas above and beyond their priorities?”
  • Would the absence of this per­son have a neg­a­tive impact on the busi­ness and on customers?”

The fact that man­agers are hav­ing reg­u­lar check-ins with their staff and are see­ing feed­back from a range of sources through­out the year also helps to increase objec­tiv­i­ty. Addi­tion­al­ly, HR busi­ness part­ners are able to chal­lenge deci­sions where they feel there is unfair bias.

So far, the results at GE have been incred­i­bly pos­i­tive with employ­ee sat­is­fac­tion regard­ing pay awards now high­er than when rat­ings were used and 77% of man­agers say they find it eas­i­er to dif­fer­en­ti­ate per­for­mance under the new approach.

The sim­pli­fied rat­ing approach

A num­ber of organ­i­sa­tions, includ­ing some of our own cus­tomers, hav­ing recog­nised that tra­di­tion­al 4 or 5 point rat­ings are actu­al­ly hard for man­agers to use objec­tive­ly and con­sis­tent­ly, have opt­ed for a sim­pli­fied approach where­by employ­ees are either des­ig­nat­ed as On track’ or Off track’, or sim­i­lar ter­mi­nol­o­gy. Where an employ­ee is On track’ they receive a pay or bonus award, based on one or more of the following:

  • com­pa­ny performance
  • divi­sion­al performance
  • team per­for­mance
  • mar­ket rate
  • changes in job responsibilities

Where an employ­ee is off track’, their pay or bonus award is reduced.

This approach has a num­ber of advan­tages. It is sim­ple so every­one under­stands it, it saves time as there is no cal­i­bra­tion required, it reduces anx­i­ety amongst employ­ees as most employ­ees will be on track’ most of the time, yet it still allows under-per­for­mance to be addressed. Addi­tion­al­ly, as man­agers only have two rat­ing options, they are able to rate employ­ees much more objec­tive­ly than with a mul­ti-point rat­ing scale, even if they are not expe­ri­enced per­for­mance raters.

The major dis­ad­van­tage of this approach is that it has no scope to award extra pay to high­er performers.

The par­tial dif­fer­en­ti­a­tion approach

The pur­pose of most per­for­mance relat­ed pay schemes that I have encoun­tered is to recog­nise high per­form­ers with addi­tion­al reward and to pay less to poor per­form­ers. If that is the aim, then we don’t actu­al­ly need to dif­fer­en­ti­ate every sin­gle employee’s pay and per­for­mance. Instead, we can sim­ply iden­ti­fy just our high­est and low­est per­form­ers to dif­fer­en­ti­ate their pay and give the major­i­ty of employ­ees who are on track the stan­dard pay or bonus per­cent­age based on com­pa­ny or divi­sion­al per­for­mance. This approach is rec­om­mend­ed by McK­in­sey who state that many com­pa­nies now think it’s a fool’s errand to iden­ti­fy and quan­ti­fy shades of dif­fer­en­tial per­for­mance among the major­i­ty of employ­ees, who do a good job but are not among the few stars.”

I per­son­al­ly like this approach a lot as it sim­pli­fies things sig­nif­i­cant­ly. We need to recog­nise that how­ev­er we decide to mea­sure per­for­mance, there will always be fur­ther dis­cus­sion and analy­sis that needs to take place to help ensure fair­ness and to deal with the inevitable dis­agree­ments that arise. So if we can lim­it the scope of this to say just the top and bot­tom 10 – 20% of the work­force, we’re sav­ing a sig­nif­i­cant amount of time, effort and poten­tial staff dis­sat­is­fac­tion. We’re also vast­ly lim­it­ing the scope of poten­tial subjectivity.

So how do you go about iden­ti­fy­ing your top and bot­tom per­form­ers with­out a rat­ing? It may be tempt­ing to put togeth­er some kind of pseu­do-sci­en­tif­ic for­mu­la based on weight­ed achieve­ment of objec­tives and demon­stra­tion of val­ues or behav­iours, some­thing that many per­for­mance man­age­ment sys­tems offer (not ours). How­ev­er, these sys­tems end up being coun­ter­pro­duc­tive because:

  • They are time con­sum­ing to com­plete so they end up becom­ing a tick-box’ exer­cise, detract­ing from mean­ing­ful, per­for­mance enhanc­ing discussions
  • Employ­ees need to under­stand the for­mu­las and weight­ings behind them in order for them to form an opin­ion about whether or not they have been treat­ed fair­ly, some­thing that is hard and time con­sum­ing to achieve
  • Once employ­ees and man­agers under­stand the for­mu­las, they will inevitably play the sys­tem’ to get the best reward out­come. Research has shown that when employ­ees are assessed against achieve­ment of spe­cif­ic objec­tives, they avoid set­ting stretch­ing objec­tives and over­all per­for­mance ends up decreasing.
  • There are many fac­tors out­side of the employee’s con­trol that impact their abil­i­ty to achieve their objec­tives. So reward­ing them against achieve­ment of objec­tives ends up either demo­ti­vat­ing employ­ees, or with man­agers argu­ing that excep­tions need to be made for spe­cif­ic employ­ees (mean­ing that we are effec­tive­ly back to man­ag­er dis­cre­tion again but with a lot more work involved!)

The solu­tion is once again to keep things as sim­ple as pos­si­ble. If you want to find out who your best and poor­est per­form­ers are, ask your man­agers. But to ensure objec­tiv­i­ty, you need to ask the right ques­tions, which leads us onto our final per­for­mance mea­sure­ment approach of using tar­get­ed questions.

The tar­get­ed ques­tions approach

In 2015, Deloitte worked out that they were spend­ing 2 mil­lion hours a year rat­ing every employee’s per­for­mance and poten­tial, col­lat­ing those rat­ings, dis­cussing and cal­i­brat­ing them. They also realised that the end results were not par­tic­u­lar­ly accu­rate, mean­ing all that time and effort was achiev­ing very lit­tle. So instead of rat­ing employ­ees as part of an annu­al appraisal, their man­agers now answer four ques­tions about each of their team mem­bers each quarter:

  1. Giv­en what I know of this person’s per­for­mance, and if it were my mon­ey, I would award this per­son the high­est pos­si­ble com­pen­sa­tion increase and bonus.
  2. Giv­en what I know of this person’s per­for­mance, I would always want him or her on my team.
  3. This per­son is at risk for low performance.
  4. This per­son is ready for pro­mo­tion today.

Whilst there will always be an ele­ment of sub­jec­tiv­i­ty when mea­sur­ing per­for­mance, it’s easy to see how these ques­tions are more like­ly to yield more accu­rate results than a clas­sic 1 – 5 rating.

Although I like the con­cept of these ques­tions, they are par­tic­u­lar­ly suit­ed to Deloitte’s pro­fes­sion­al ser­vices cul­ture. So I spent some time devel­op­ing a set of ques­tions that could be used in a wide range of organ­i­sa­tions who wish to:

  • iden­ti­fy and pro­vide addi­tion­al reward to their top per­form­ers (a good strat­e­gy accord­ing to McK­in­sey who found that 80% of val­ue is gen­er­at­ed by only 20% of employees)
  • iden­ti­fy poor per­form­ers (and poten­tial­ly give them less reward)
  • iden­ti­fy employ­ees with poten­tial for pro­mo­tion right now

Here are the 5 ques­tions that I came up with:

  1. Has this per­son made an EXCEP­TION­AL impact above and beyond their job descrip­tion and agreed objec­tives through­out the year? [Yes/​No]
  2. Has this per­son CON­SIS­TENT­LY demon­strat­ed our organisation’s val­ues and behav­iours through­out the year? [Yes/​No]
  3. If you have answered YES to BOTH of the above two ques­tions, are you nom­i­nat­ing this per­son to receive an addi­tion­al dis­cre­tionary bonus? [Yes/​No]
  4. Do you have con­cerns about this person’s per­for­mance over the last 6 – 12 months? [Yes/​No]
  5. Is this per­son ready to be pro­mot­ed today? [Yes/​No]
    Where you have answered YES to any of the above ques­tions, please pro­vide addi­tion­al details to sup­port your answer(s)

Ques­tions like this can eas­i­ly be added into the View­points mod­ule of our Clear Review Per­for­mance Man­age­ment soft­ware, enabling you to col­late the answers from man­agers online and analyse them quick­ly and eas­i­ly. Here’s what a man­ag­er would see when answer­ing these ques­tions in Clear Review:

Performance related pay questions on Clear Review