The Human Resources Institute of New Zealand - Logo

The Human Resources Institute of New Zealand

Human Resources Institute of New Zealand (HRINZ) is the professional body for those involved in Human Resource Management and the development of people.

HRINZ represents the interests of 3,000+ individual members who make up around 45% of the known New Zealand HR market. Read More

Profile ID

Password

Remember me


Performance Related Pay Review - Using a Matrix

Deciding how to use the funds available for a remuneration review is always an interesting challenge, especially where the money available is tight and needs to be used smartly. Most organisations require their managers to make decisions, usually around the outcomes of performance reviews. This is not particularly popular with some managers, but it is a part of the job! If the organisation takes performance seriously, then linking it to the remuneration review is a valid action. It reinforces the messages that employees should be receiving about their contribution to the business. The money may or may not change behaviours – but it helps to make the point, by restating and reinforcing the values espoused by the organisation.

That said, the performance review is NOT about pay. It is about performance, and must not be hi-jacked or distracted. The pay review should recognise its own rationale, and its logic and process need to be clearly understood by those responsible for it.

This note is designed to explain some of the thinking that goes into setting up a simple mechanism for running a performance-related pay review, and the use of a matrix to that end.

Why review pay at all?

If an employer wants a pay structure that supports the hiring and retention of the sort of staff needed to achieve the business goals, that structure needs to be reviewed periodically to ensure that it continues to be up-to-date. Otherwise, there will be a growing regularity of paying above company policy, which leads to inequity. The employment market changes over time, and different parts move at different speeds, so updating the pay ranges for staff helps to ensure that new staff are able to be recruited at market-relevant pay levels. This is an administrative first step, simple system maintenance, and does not of itself require change to existing pay levels.

Retention of existing capable staff then becomes the issue. The market value of a role will probably have changed – the market updates to the ranges will show roughly how much - so keeping pace with the market entails looking at current staff’s pay.

Some organisations simply pass on to staff the amount by which their ranges have moved, but I believe this misses an important point. The employer may be willing and able to pay more, but the rationale for any increase needs to be clear. There needs to be at least some sort of identifiable threshold of acceptable contribution, to justify the increase. Otherwise, pay increases become an expectation and a perceived entitlement for simply still being in the job as time passes.

This is an opportunity for an organisation to restate its values, and it is sensible to use the leverage of pay increases to reinforce those messages to the managers and staff.

Performance is the commonly stated reason for increases, whether this be through measured performance or the more indirect ‘growth of capability’. The obvious corollary to the latter is that the capability is relevant to the role, and actually put into practice. Otherwise the organisation is paying staff to learn someone else’s job, quite possibly with their next employer!

An organisation may in reality use the pay review funds to reward something as well as, or instead of, performance; and that’s o.k. as long as the ‘something else’ is as important as performance in the organisation’s eyes. The mantra of performance-related pay reviews is that:-

If you’re not rewarding performance, you’re rewarding something else.

If the something else is that important, it should feature in the organisation’s remuneration strategy, and have senior management blessing. If it doesn’t, then it shouldn’t have a claim on the pay review dollars.

So why not just have the same percentage increase for the same performance rating?

This would be understandable if everyone were paid the same, but usually they aren’t. If the performance is the same but the pay is different, giving the same increase to all will simply compound the existing (and no longer justified) differences. Good performers at lower pay are more vulnerable to feeling aggrieved, leading to demotivation and lower contribution - or departure.

So it makes sense to gear any review dollars towards staff who have performed better than their current position in pay range rewards them.

Who decides the level of increases?

The larger the organisation, the harder it is to achieve consistency across different operational areas. It can be achieved if the allocation is determined centrally according to standardised criteria, but this suffers the major drawback of taking the responsibility for pay considerations right out of the hands of line managers, where at least some of the responsibility belongs. Good managers feel disempowered: poor managers feel relieved, and disclaim responsibility, blaming HR and ‘the system’. You may well have heard “If it was down to me ….”.

If however the whole review process is left as a lolly scramble, with the most aggressive managers grabbing the most, inequity is inevitable, favouritism and internal poaching will flourish, and the organisation will be buying grief.

One way of resolving this is through the use of what I call the ‘tolerance matrix’.

When is a matrix not a matrix?

It is a relatively common practice nowadays in pay reviews to use a “performance pay matrix”. It is a useful tool in the review process, and quite easy to construct. Some managers have said to me ‘We don’t use a matrix in our organisation’. Further enquiry showed that they did – they took into account individual performance and current pay levels in determining the pay review distribution – they just hadn’t drawn the pictures. They had a ‘virtual matrix’!

A performance-related matrix is a framework that guides managers in deciding how remuneration review dollars might be allocated. It relies on two factors or co-ordinates – performance rating and position in range (or PIR). It does not have to be definitive in setting review percentages, and can be used simply as a preliminary guide, but it can help to ensure a consistency of approach on the part of different managers and operational areas, leading to a more equitable spread of the available funds. It also makes it easier for management to align overall review increases with the available funds.

The two standard co-ordinates are:-

a) Relative remuneration position (PIR) and

b) Performance measurement

There are also secondary co-ordinates of

c) available funds and

d) Results gearing

And if you want to be really cool, you can look at e) recommendation tolerances.

The shape of the matrix can remain unchanged from year to year, unless the remuneration and/or performance measurement systems change materially.

The numbers/percentages that appear in the boxes will need to be revisited in advance of each remuneration review, because they reflect the funds available and the company’s intentions in respect of distribution.

Some users have struggled in the past because they have tried to use the same percentage figures year on year, but found that the resulting sum of increases is quite out of line with the funds available – sometimes leaving a lot to spare, but more commonly blowing the budget. The matrix gets the blame, but the real problem lies with using it in an inflexible way.

Drawing the boxes actually helps, because it offers a logically-constructed model of the though process, built on logic, drawn on paper, shared with others and driving consistency.

That said, there are many ways that a matrix can be constructed.

What does a matrix look like?

Performance

Performance doesn’t matter

PIR

Paid anywhere inside or outside range

X%

The simplest is the least satisfying from the performance point of view. It covers “cost-of-living” type increases. It gives everyone the same percentage, irrespective of their performance or where they stand in their range.

The critical point is that performance is clearly stated as not important enough to make a difference. Just show up reasonably frequently and you get a pay rise.

Interestingly, as an employee and as a consultant over many years, I have found this is less popular with staff than with managers. Some managers like it because it is easy and they can avoid making decisions. Many staff dislike it because they see incompetence being rewarded: the people they have carried for the year get the same as themselves.

Performance

Good

Indifferent

Bad

PIR

103 – 110%

0

0

0

98 – 103%

X-

0

0

92 – 97%

X

0

0

87 – 91%

X+

X-

0

80 – 86%

X++

X

0

The next stage for greater sophistication comes by sub-dividing the categories, differentiating on the basis of the key co-ordinates - performance and current pay level.

The model here assumes:-

· a 3-step performance scale;

· 80% to 110% ranges;

· a review pool of around X%;

· an intention to reward staff higher for good performance if they are low in their range – i.e. moving them to a better PIR; and

· a willingness to offer zero to a proportion of staff.

So of course, a matrix needs to be drawn to reflect each organisation’s own idiosyncratic systems of range setting and performance measurement.

There are two important themes in such a model:-

a) This is not the place to reward poor performers. Managers often use review funds to reward poor performers for other things they value – length of service, loyalty, enthusiasm or just being there. The company may value these too: if so there should be other, more pertinent, avenues for their recognition and reward. But remember:-

If you’re not rewarding performance you’re rewarding something else.

b) Good performance is rewarded in context. Payment high in the range places a higher expectation on performance. If someone is paid at 110% of their new midpoint, they are already being rewarded in advance for outstanding performance of their role, and that performance level has to have been sustained just to merit staying there. Whilst the employer may be averse to moving pay downwards, there is no reason for it to increase. [There may be scope for a one-off payment if performance is high and funds permit.]

Conversely, someone paid low in their range merits a higher than normal increase for a high level of sustained performance, moving them closer to a position in the new range that corresponds with their level of contribution.

The setting of different levels of reward in different boxes of the matrix is designed to set pay according to realistic performance expectation for the coming period.

Gearing the funds

Gearing the distribution of the available funds for the pay review entails a number of estimates:-

[Where the average/median performer stands (the shaded box here);

[The funds available for the review, as a percentage of the funds reviewed (4% here).

Performance

Super

Very good

Fair

Poor

PIR

Over 100%

0

0

0

0

95 – 100%

4

2

0

0

90 – 95%

6

4

2

0

85 – 90%

8

6

4

0

In general terms, and where the intention is to spread the funds fairly according to performance and PIR, the overall review percentage would appear in the shaded box.

This “overall percentage for the average result” would be the benchmark against which the other boxes would be assessed, in a process similar to the model

– the higher the PIR, the higher the expectation/the lower the reward.

The pattern may vary where other principles apply – e.g. poor performers get nothing, even if at the bottom of their range, or no systemic increases for staff above 100%, or any other such premise.

Following a process of this nature produces a result roughly in the region of the funds available: the high increases merited in the bottom left corner are offset by the zeroes and 2% staff.

Variations

It is generally accepted that managers should be involved in, and take some responsibility for, the remuneration of their staff. A common complaint against using a matrix in a performance-related remuneration review is that it is too prescriptive, and offers no flexibility – and where managers cannot influence pay by their recommendations, they do so indirectly by adjusting the performance markings.

Permitting managers to override the recommended levels is one option, though this can bring the whole process into question if used inconsistently across divisions, and therefore requires protocols for exercising discretion.

Performance

Super

Very good

Good

Fair

Poor

PIR

Over 100%

0 – 1%

0-1

0

0

0

95 – 100%

2 - 5

1 - 3

1 - 2

0

0

90 – 95%

4 - 6

2 - 5

1.5 – 4.5

0 – 1.5

0

85 – 90%

5 - 8

4 - 6

2 - 5

0 - 2

0

80 – 85%

8 - 10

5 - 8

3 - 8

2 - 4

0

 

 

 

 

Another option is the use of the “tolerance” matrix, such as the table below. This offers limited percentage review ‘mini-ranges’, so managers have a degree of discretion and control over the pay of their staff.

In this model, the value 3% represents the percentage of funds available for the review, and the shaded box again represents the most common mix of performance marking and PIR.

Assuming a reasonable use of the tolerances, the overall results are likely to be similar to those of the more prescriptive model.

And as with any review process, recommendations outside the tolerances could be submitted for the CEO’s/General Manager’s consideration.

Some common questions

Does one matrix cover a whole organisation, or should each division/business unit have its own?

The more that can be covered by one matrix, the better for cost management and application consistency. One matrix can cover the whole organisation as long as the same principles apply to all, i.e. the spread of pay ranges (80%-110%, or whatever); the number of performance ratings; the size of the review pool. Minor variations can be managed within the single framework: any larger differences may require a degree of customisation.

Does a one-size percentage fit all in a performance rating category? The new range midpoints have moved at different speeds.

In updating the range midpoints for the review, it will be noted that the new midpoints will usually have changed by different percentages since last time. This is simply the market effect.

The underlying principle is that the new midpoint is the sole reference point for the coming year, and only comparison against that is relevant.

In practice, the PIR for remuneration review purposes has been assessed on the new midpoint. If the midpoint has moved a long way since last year, then the employee’s new PIR will be that much lower, and they may qualify for consideration of a larger increase.

We do not use performance ratings. Can we use a matrix?

Not as such. A matrix needs a performance-related co-ordinate, to function. The rating may be only an over-simplified plastic ‘widget’ to link the performance review to the pay review, but it is an essential part of the mechanism. After all, if it is a performance-related pay review, there has to be some sort of criterion for gearing the performance measurement to the pay outcome.

It may be feasible to devise a temporary rating system that is sufficient for the purpose, to the extent of delivering suggested percentage outcomes for managers to consider as a first stage in the review process.

But if you are determined to have no overall rating system, then a performance-related matrix is not for you.

We have different range percentage spreads for different parts of the organisation. Can we use a single matrix?

The strength of the single matrix is that it can cover all levels of job, but its statistical validity depends on the same rules applying to everybody covered. If the differences are material (e.g. one area is 75%-100% and another is 95%-125%) a single matrix will not work properly. You may need to have two (or more).

How do we split the cost-of-living/range increase element from that related to performance?

You shouldn’t, in my view! All of the increase is performance-related, in the sense that there should be no increase without satisfactory performance. The market data that you have used to update the ranges will encompass all the market factors that have influenced pay movement; so paying range movement plus performance is, at least in part, rewarding twice for the same thing.

The updated ranges represent the new market against which your organisation is setting and reviewing its pay practices. The issue is one of “This is the market for X’s job. He/she should probably be paid around Y% of their midpoint, and I will try to make it so, if Saveaffordable”. There need be no distinction as to what makes it up: the dollars all look the same, and the goal is to pay someone wherever in their range they deserve to be, subject to what the organisation can afford. This is a different mindset from the ‘status quo, plus’ thinking that is driven by factors other than the remuneration market data to which the organisation’s remuneration strategy presumably subscribes.

Summary

This is of necessity a condensed explanation of the performance-related matrix: because every organisation is different, the degree of customisation to the basic model will differ to some degree. [Your friendly neighbourhood remuneration advisor/consultant can colour in the variables, if you are unsure of the process.]

A performance-related matrix is not a magic wand: it does not invent money where it does not exist, or allow managers to avoid making hard decisions.

But it is a simple and flexible tool, capable of customisation to many different scenarios, and can offer a high degree of operational consistency, equitable process and cost management, whilst giving line managers pay-setting responsibility within reasonable parameters.

HRINZ would like to express their gratitude to  Alan White, Remuneration Consultant, for writting this article especially for publication on this specialist area, within the HRINZ Knowledge Base.

 


Knowledge Base Search | Browse Topics: A-H  I-P  Q-Z |

Can’t find what you’re looking for? Contact HRINZ for help.

For articles, tips and tools members can view HR Guides, Human Resources Magazine and Other HR Articles.

Disclaimer: This information has been written for and submitted to HRINZ for publication and has been published in good faith for the general information of HRINZ Members of the Institute. HRINZ accepts no legal responsibility for the contents of the Knowledge Base and appropriate professional advice and assistance should be sought in particular cases.

MoST Content Management V3.0.6894