A merit increase is a salary raise given to recognize individual performance. It's the main mechanism most companies use to differentiate pay based on contribution — and also one of the most frequently botched processes in HR.
Done right, merit increases reinforce your performance culture: employees who contribute more earn more, and the connection between contribution and reward is clear. Done wrong, merit increases are just cost-of-living adjustments with extra paperwork — everyone gets roughly the same raise, managers hate the process, and high performers start looking elsewhere.
This guide covers the mechanics of merit increases: how to design the matrix, how to connect merit to calibrated performance ratings, and how to communicate increases in ways that actually motivate people.
What you'll get: A merit increase matrix template, a framework for connecting merit to performance ratings, a budget modeling approach, and communication scripts for managers.
Merit Increases vs. Cost-of-Living Adjustments
Before building your merit framework, get clear on the distinction between merit increases and cost-of-living adjustments (COLAs). They serve different purposes and should be treated as separate budget items.
| Merit Increase | Cost-of-Living Adjustment (COLA) | |
|---|---|---|
| Purpose | Recognize individual performance and contribution | Maintain purchasing power against inflation |
| Basis | Performance rating, compa-ratio, time in role | CPI / inflation rate or market salary movement |
| Who gets it | Employees whose performance justifies it | All employees (or all above a threshold) |
| Differentiation | High variation (e.g., 0%–10% based on performance) | Uniform (e.g., 2% for everyone) |
| Communication | "Your raise reflects your strong performance this year" | "We're adjusting all salaries to keep pace with market" |
In practice, most organizations blend the two into a single annual increase. The problem: when a 3% raise is actually 2% COLA + 1% merit, neither the retention value nor the recognition value is very strong. Separating them — at least in how you communicate them — makes each more meaningful.
Designing the Merit Increase Matrix
A merit increase matrix connects performance ratings to merit increases while also accounting for where an employee sits in their pay range (compa-ratio). The result is a two-dimensional grid that gives managers clear guidance without removing all judgment.
Why compa-ratio matters
An employee at the bottom of their pay range (compa-ratio 80%) has more room to grow and more retention risk if left underpaid. An employee at the top of their range (compa-ratio 120%) is already well above market — giving them another 5% raise just moves them further above the max and signals that your ranges don't mean anything.
The matrix accounts for this automatically.
Sample Merit Increase Matrix (3.5% Total Budget)
| Performance Rating | Compa-Ratio <80% | Compa-Ratio 80%–95% | Compa-Ratio 95%–110% | Compa-Ratio >110% |
|---|---|---|---|---|
| Exceptional (top 10%) | 9%–12% | 7%–9% | 5%–7% | 3%–5% |
| Exceeds Expectations (top 25%) | 6%–8% | 5%–7% | 3%–5% | 1%–3% |
| Meets Expectations (middle 50%) | 4%–5% | 3%–4% | 2%–3% | 0%–2% |
| Partially Meets (15%) | 1%–2% | 0%–1% | 0% | 0% |
| Does Not Meet (bottom 5%) | 0% | 0% | 0% | 0% |
Calibrating the matrix to your budget
The specific percentages in your matrix depend on your budget and workforce distribution. Here's how to build the matrix from your budget:
- Start with your total merit budget (e.g., 3.5% of total base payroll)
- Estimate the distribution of your employees across performance rating and compa-ratio buckets
- Plug in provisional increase amounts and calculate the weighted average cost
- Adjust until the weighted average cost equals your budget
Example: If 65% of employees are "Meets Expectations" and 15% are "Exceeds," you need those large buckets to average close to your budget rate. Small buckets (exceptional performers, low compa-ratio) can be more generous without blowing the budget.
The Role of Performance Calibration
A merit matrix is only as good as the performance ratings feeding into it. If managers give everyone "Exceeds Expectations" to justify higher raises for people they like, the matrix becomes meaningless — and your merit budget ends up distributed by manager generosity, not employee performance.
Performance calibration solves this. Calibration is the process of reviewing and normalizing performance ratings across teams before they're finalized — ensuring that "Exceeds Expectations" means the same thing in Engineering as it does in Marketing as it does in Sales.
The connection to merit increases:
- Calibrated ratings produce merit distributions that reflect actual performance across the organization, not individual manager bias
- Calibration surfaces the relative ranking of employees, which helps managers defend merit decisions to their teams
- When employees know their rating was calibrated against peers — not just handed out by their manager — the merit increase feels more legitimate
Without calibration, merit increases feel random. With calibration, there's a defensible connection between what someone contributed and what they earned.
Confirm's performance management platform includes built-in calibration tools that connect directly to compensation planning — managers see each employee's performance rating alongside their compa-ratio, making merit increase decisions faster and more consistent.
Budget Modeling Before the Cycle Starts
Before you hand managers their merit budgets, model what different budget levels will cost and what outcomes they'll produce. This prevents two common problems: under-budgeting (managers end up with less than they need to retain key people) and over-spending (HR ends up chasing managers down to reduce their worksheets after the fact).
A basic model:
| Scenario | Budget % | High Performer Increase | Mid Performer Increase | Annual Cost |
|---|---|---|---|---|
| Conservative | 2.5% | 4%–6% | 1%–3% | $[calculate] |
| Base case | 3.5% | 6%–9% | 2%–4% | $[calculate] |
| Competitive | 5.0% | 8%–12% | 3%–5% | $[calculate] |
Run the model across your actual population distribution before presenting scenarios to leadership. "3.5% budget" sounds abstract. "$1.4M in additional salary cost producing an average 6% raise for top performers" is a decision leadership can actually evaluate.
When Merit Increases Aren't Enough: Off-Cycle Adjustments
The annual merit cycle can't solve every compensation problem. Situations that require off-cycle adjustments:
- Promotion — when someone moves to a new level, their salary needs to move to the new range immediately, not wait for the next cycle
- Market correction — when market data shows a specific role is significantly below market due to rapid salary inflation (this happened with software engineers in 2021–2022)
- Retention counter-offer — when a top performer has an outside offer that exceeds their current comp range; better to adjust proactively than react to a resignation
- Pay equity remediation — when an audit finds a pay gap that needs to be corrected before the next cycle
Off-cycle adjustments should be the exception, not the norm. A well-designed merit process reduces the need for them by keeping pay current annually. But having a clear policy for when they're approved — and who approves them — prevents ad-hoc decisions that create new equity problems.
Communicating Merit Increases to Employees
The delivery matters as much as the amount. A 5% raise communicated poorly feels like 2%. A 2% raise communicated clearly — with the reasoning — lands better than you'd expect.
What managers need to communicate:
- The performance summary — what the employee accomplished that's being recognized
- The specific increase — the new salary and the increase amount/percentage
- The context — where they sit in the pay range and how the increase affects that position
- The forward-looking piece — what continued strong performance leads to
Sample manager script (high performer):
"I want to walk through your compensation for this year. Based on your performance — [specific achievements] — you've been rated Exceeds Expectations in calibration. That's in the top 20% of your peer group.
Your salary is moving from $[X] to $[Y], an increase of [Z]%. That puts you at [compa-ratio]% of our market midpoint for your role and level — in the upper portion of the range, which reflects where your performance has you.
I want to be transparent: the merit budget sets a ceiling, and I used as much of my budget as I could to recognize your contribution. If you have questions about the process or where the budget comes from, I'm happy to walk through it."
What not to say:
- "This is the best I could do" — undermines confidence without giving useful context
- "Everyone got similar increases" — if true, suggests your merit process doesn't differentiate (fix that); if untrue, don't say it
- "I fought for you" — implies the process was arbitrary and personality-based
- "The company is going through tough times" — if you're giving merit increases at all, this framing contradicts itself
Merit Increases and Pay Equity
Merit increases are a major driver of pay gaps. If managers consistently give higher merit increases to employees of certain demographics — even unconsciously — the gaps compound over time. A 1% gap in merit increases becomes a 10%+ pay gap over a decade.
Before finalizing merit increases, run this check:
- Pull the proposed merit increases by demographic group (gender, race/ethnicity)
- Control for performance rating and compa-ratio (employees with the same rating and similar compa-ratio should get similar increases)
- If any group is getting systematically lower increases after controlling for those factors, stop and investigate before distributing increases
This is not a complex statistical analysis — it's a basic check that takes 30 minutes in Excel. It's one of the highest-value things HR can do during the merit cycle. See the full pay equity audit framework in the Pay Equity Guide.
Frequently Asked Questions
How do I handle merit increases for employees who are already at the top of their range?
Employees at or above the range maximum ("red-circled") typically receive 0% merit increase and instead receive a one-time lump sum bonus equal to what their merit increase would have been. This recognizes their performance without permanently increasing salary beyond the range. If someone is consistently performing above their range maximum, review whether the range is correct or whether a reclassification is warranted.
What's the right merit budget for 2026?
Most compensation surveys project 3.2%–3.5% as the median merit budget for 2026. Companies in competitive talent markets (tech, healthcare, finance) typically run 4%–5% or higher. Budget at the median and model scenarios — knowing what a 4% budget produces in high-performer increases vs. a 3% budget helps leadership make an informed tradeoff.
How do we prevent managers from using their full budget on mediocre performers?
Three mechanisms: (1) Hard-coded matrix guidance (or ranges) in the manager worksheet that limits how much can go to employees below a certain rating; (2) calibrated performance ratings before managers see the budget, so they can't retroactively inflate ratings to justify larger increases; (3) HR review of manager worksheets before approval, with flagging of outliers. None of these are perfect, but together they prevent the worst patterns.
Should new hires get merit increases in their first year?
Most companies exclude employees who have been in role for fewer than 6 months from the merit cycle. Employees hired at 7–12 months may receive a prorated increase or may be deferred to the next cycle, depending on how their offer was positioned relative to the range. The policy should be consistent and communicated at hire — nothing surprises a new employee more than being left out of their first annual review because of an unstated eligibility rule.
