Table of contents
NOTE: Table of contents generated on published site only, does not display here. If no H2s are present in the article, the TOC should be turned off in the article colleciton entry.
Share this content

A single year of merit cycle data reflects a moment in time. Three years uncovers a trend.

This year marks the third consecutive edition of Pave's Merit Cycle State of the Union, and now we can connect the dots across three years of raises, promotions, and performance decisions. Drawn from a dataset covering 200+ companies and more than 100,000 employees, we’ve identified the patterns shaping merit cycles.

Here's what three years of data shows us.

  • Merit freeze rates have leveled off
  • Budgets are slowly shrinking
  • Promotion rates are holding steady
  • Pay for performance is intact, but equity is doing the heavy lifting

1. Merit Freeze Rates Have Leveled Off

When cash is tight, some companies skip merit cycles altogether—a practice known as a merit freeze, in which few employees receive a salary increase. When faced with a choice between layoffs and forgoing merit cycles, many organizations choose the latter to preserve cash without reducing headcount.

The good news: the freeze era appears to be over. The sharp drop in freeze rates that defined 2024 has leveled off, with behavior stable in the trailing 12 months through April 2026. 

“Some companies were stuck between a rock and a hard place. They could lay off people or freeze merit cycles to save payroll costs. Now, when I look at the data, I see, for the most part, that the skip the merit cycle behavior that companies were trying a few years ago is mostly over.”
—Matt Schulman, CEO and Founder at Pave

Hear more from Matt—view the on-demand Merit Cycle State of the Union webinar.

That said, smaller companies (101–200 employees) are still freezing at three to four times the rate of their enterprise counterparts—a pattern consistent across all three years of data. For those companies, freezes remain a financial necessity driven by runway preservation and capital efficiency, not compensation philosophy

For companies still navigating freeze constraints, the most effective approach has been shifting focus to equity and non-cash levers, like career development, leadership visibility, and flexibility, to retain and motivate top talent without increasing spend.

2. Budgets Are Slowly Shrinking 

Merit budgets aren’t bouncing back. The median salary increase among eligible employees has trended downward three consecutive years—from 3.6% in H1 2024 to 3.5% in H1 2025 to 3.4% in H1 2026. Across employees who received a raise, the median dropped from 5.2% to 4.9% over the same period.

However, it’s important to note that the distributions are fairly wide, with some companies landing below the median and others above the 75th percentile.

With reduced budgets, the question has become: how do companies allocate their money most effectively?

{{mid-cta}}

3. Promotion Rates Are Holding Steady 

Overall promotion rates have held steady at 8.6% across all employees in both H1 2025 and H1 2026, roughly flat with H1 2024. 

But more nuance exists when comparing promotions by company size and function.

  • Company size: Smaller companies have historically promoted at higher rates than their enterprise counterparts, but that gap is closing quickly. The 1–200 employee cohort dropped from 12.5% to 10.7% in a single year, almost converging with larger company cohorts that hover between 9–11%.

  • Function: R&D received the smallest promotion raise of any department in H1 2026 at 7.7%, down from 9.9% in 2025. But companies aren’t underinvesting in technical talent; they’re just using other levers, like annual raises, to retain those employees. 

4. Pay for Performance Is Intact, but Equity Is Doing the Heavy Lifting

Three years of data show a consistent pay-for-performance structure: performance rating affects raise amounts, equity participation, and grant size. What's also consistent is that cash isn’t the only lever for rewarding top performers.

Above and Meets Expectations employees receive raises at similar rates: 87.5% and 84.4% respectively in H1 2026. But as Devin Engelsen, Managing Principal at Wailea Partners, noted during the 2026 Merit Cycle State of the Union webinar, the difference between rewarding top performers and average performers goes beyond salary: "From what I see, the differentiation is on equity."

Equity participation drops significantly between Above Expectations (47.5%) and Meets Expectations (25.4%)—a gap far wider than anything in the cash data. Meanwhile, Below Expectations’ equity participation has effectively reached zero in 2026. 

Companies are implementing pay for performance across the board, but the biggest difference between top and low performers often comes down to equity participation and grants.

Run Merit Cycles With Confidence

Three years of data point to a clear shift: companies are running more disciplined and deliberate cycles, and making thoughtful choices about who gets raises, who gets promoted, and where equity is granted. In a market where budgets aren't expanding, the quality of those decisions matters more than ever.

With access to the right data and tools, compensation leaders can move from reactive to intentional, building merit cycles that retain and motivate top talent.  

Download the 2026 Merit Cycle State of the Union for more insights, including job family-level raise data, pay-for-performance benchmarks, and full geographic breakdowns.

Share this content

Pave is a world-class team committed to unlocking a labor market built on trust. Our mission is to build confidence in every compensation decision.

NOTE: The elements below are only visible in the editor. To place these in articles, use their corresponding short codes. They are made visible here to facilitate editing.
{{mid-cta}}

Benchmark Your Merit Cycle

Get insights on merit cycle trends, with data from over 100,000 employees.
{{signup-cta}}
{{signup-cta-narrow}}
{{article-cta}}
Market Data Pro

Harness real-time benchmarks. Sync with industry standards

{{newsletter-cta}}
{{article-stats}}
No items found.
{{key-results}}
Key results