Compensation reviews are one of the most complex processes a Reward team runs. You're trying to reward strong performance, stay competitive with the market, and maintain pay equity – all at once, across an entire organisation, with a fixed budget.
Without a structured framework, something usually gives.
Managers apply increases inconsistently. High performers in already well-paid positions drift further from peers doing equal work. Pay equity issues that were already present get quietly compounded with every review cycle.
A merit matrix solves this by bringing structure to one of the hardest decisions in the process: how much should each employee's pay increase, and why?
A merit matrix is a structured framework used during compensation reviews to determine how much an employee’s pay should be adjusted. It combines two key metrics: an employee’s annual performance rating and their position within their current salary band.
Sometimes referred to as a merit grid or salary increase matrix, the primary purpose of a merit matrix is to ensure consistent and fair decisions during a pay review.
The main benefit of a merit matrix is to ensure pay increases are both performance-driven and equitable.
Employees with the same performance rating receive different increases based on where they sit in their salary band – so someone paid below the midpoint gets a higher increase than a peer at the top of the band with the same rating.
This matters because two employees doing equal work can easily end up at different salaries – through differences in starting salary negotiations, or simply through tenure and accumulated pay increases over time. Without accounting for band position, the same performance-based increase for both widens that gap further with every review cycle.
Used correctly, merit matrices also make merit increases more market-aware.
If salary bands are refreshed against up-to-date benchmarks before the matrix is applied, the band position automatically reflects how competitive an employee's pay is relative to the market target as well as relative to their peers.
"A merit matrix is an effective and transparent tool for rewarding performance. It simplifies the process, eliminates unnecessary noise, and provides the clarity that people value. While it can feel overly structured for smaller organisations, the positives – especially when supported by fair performance assessments – far outweigh the negatives. For companies committed to rewarding performance, a merit matrix is a valuable framework."
Chief People Officer at Ravio
💡 How commonly used are merit matrices?
Data from Ravio’s 2025 Compensation Trends report highlights an also reveals a near-even split in whether companies use a merit matrix to determine their pay increases – with 49% of companies using merit matrices and 51% opting for alternative methods.
A merit matrix takes the form of a grid with two factors: band position and performance rating.
Each cell in the grid contains a recommended salary increase percentage – and the combination of the two inputs (performance and band position) is what determines it.
Band position measures where an employee's current salary sits relative to their salary band, typically given as either a compa ratio or salary range penetration figure.
It’s included in the merit matrix because an employee at 0.8 of their band midpoint is being paid below the market rate for their role and level – and potentially being paid inequitably compared to peers performing the same but at a higher band position.
The matrix reflects that by assigning them a higher increase, accelerating the correction of that gap to bring consistency to salaries and pay progression across the team.
💡Tip: Refresh your salary bands against current market benchmarks before you build your matrix
Band position will always be an indicator of pay equity across the band, but it’s only a meaningful reflection of market positioning if the band midpoint and range are up-to-date with current market rates.
Refreshing your benchmarks and bands first means market movements are already absorbed into the band position by the time you implement the matrix – so the matrix handles market adjustments, pay equity, and performance-based increases in one step.
Performance rating sits on the other axis, using a standardised scale (such as 1-5) that’s applied consistently across the organisation.
The higher the rating, the higher the recommended increase – but always in combination with band position, not in isolation.
This is why having a well-defined, calibrated performance review process in place before applying a merit matrix is a must. If ratings aren't consistent across teams and managers, the matrix can't do its job fairly – you'd be applying a structured framework on top of inconsistent inputs.
The combination of the two factors means that the cells where low band position meets high performance carry the largest increases – these are employees who are performing strongly but whose pay hasn't kept pace.
The cells where high band position meets lower performance carry the smallest increases, or zero. Every other combination sits somewhere in between, following the same underlying logic.
💡Tip: Once you have your matrix logic, run a test scenario before you commit to it
Apply the percentages to your actual employee population and see what the total cost looks like against your budget before rolling it out.
You might need to iterate on the cell percentages to ensure you stay within budget – but it's much easier to do that at this stage than after increases have been communicated.
Below is an example merit matrix using a 1-5 performance rating scale and compa ratios ranging from 0.8 to 1.2, with a maximum increase of 14%.
Compa ratio | Rating 1 | Rating 2 | Rating 3 | Rating 4 | Rating 5 |
|---|---|---|---|---|---|
1.2 | 0% | 0% | 2% | 3% | 5% |
1.1 | 0% | 1% | 3% | 5% | 7% |
1.0 | 0% | 2% | 4% | 7% | 10% |
0.9 | 0% | 3% | 6% | 9% | 12% |
0.8 | 0% | 4% | 8% | 11% | 14% |
To see how this plays out in practice, let’s take two employees, both rated 4 out of 5 in their performance review.
Employee A has a compa ratio of 0.85, meaning their current salary sits below the midpoint of their band. According to the matrix, they'd receive an 11% increase. Their strong performance is recognised, and the increase also begins to close the gap between their current pay and the market midpoint.
Employee B has a compa ratio of 1.15, meaning they're already paid above the midpoint. Same performance rating, but the matrix recommends a 5% increase. Their performance is still rewarded, but the smaller increase reflects the fact that their pay is already competitive within the band.
While merit matrices are a widely used method for determining pay adjustments, they’re not the only approach.
Depending on your organisation’s needs, compensation philosophy, and focus areas, there are several alternatives to consider.
Market-based adjustments focus solely on aligning employee salaries with external market rates – prioritising market competitiveness and fairness across roles and locations without considering individual performance.
Organisations relying on market-based adjustments use up-to-date salary data (through real-time salary benchmarking tools) to review and revise salaries.
Performance can still be rewarded separately through bonuses or other incentives like accelerated promotion timelines.
In this approach, managers are given a budget and broad guidelines by HR but have full discretion to allocate salary increases based on their assessment of their direct reports' contributions.
This method enables tailored rewards for unique situations, critical roles, or high-performing employees who may not fit neatly into a structured model like a merit matrix.
However, it requires consistent manager training and calibration sessions to mitigate risks of bias, inconsistency, or favouritism. Therefore, this approach is best suited for organisations with strong management training processes and confident leadership capabilities.
A simpler alternative is applying uniform salary increases across employees – which could either be:
While this method simplifies the process, it may lack the nuance needed to address pay equity or retain competitiveness in the market for each role and, particularly, for top-performing employees.
The 9-box grid is similar to a merit matrix, but with the addition of including employee potential as a factor alongside performance and band placement.
The benefit of including employee potential is that talent development and succession planning are factored into the equation – rewarding not just current performance, but also leadership potential within the company, and therefore reflecting the importance of retaining that individual long-term.
According to Rob Green, Global Rewards Consultant, Founder of Darwin Total Rewards, and an Associate within the MCR Consulting group, the 9-box grid “demonstrates organisational and leadership maturity when all three elements are working together effectively.”
Image credit: Personio
A merit matrix works best when both performance and maintaining fairness are core factors in how you adjust base salaries – and when you have the foundations in place to make that fair.
That means a consistent, calibrated performance review process, up-to-date salary bands, and a clear compensation philosophy that supports pay for performance.
If those foundations aren't there yet, the matrix can't do its job properly – and the risk is that you apply a structured framework on top of inconsistent inputs, which compounds rather than corrects pay inequities.
Your monthly dose of market insights and expert perspectives
Vaso Parisinou (CPO at Ravio) and Saime Kutuklu (Director, Central Total Rewards at Delivery Hero) had an honest retro on their latest salary review cycle – the wins, the challenges, and the lessons learned.
Merit increases can backfire without proper foundations. Discover why successful merit cycles need job architecture, salary bands & more first.
The compensation structures you have at 20 people no longer work at 200. Giac Soliman and Rob Green share their advice on building a compensation review process that scales.
