Raises Are Capped Near 3.5% Again for 2027: How to Actually Ask for More

Somewhere in your company right now, a spreadsheet has a single cell with a number in it. It might say 3.5%. That cell is the entire pool of money your manager will divide among everyone they manage next year, and by the time your performance review arrives, it has usually been decided. The conversation you think of as "asking for a raise" often happens after the part that mattered.

That is worth knowing, because the number is not generous, and it is not going to be. A salary-planning survey from Mercer, now published under the Marsh name, reports that US employers are planning a 3.2% merit increase budget and a 3.5% total salary increase budget for 2027. The survey covered 1,001 organizations. Here is what that means, and what still works.

What a 3.5% budget actually means

The first thing to notice is that 3.5% is an average across a lot of very different employers. The survey includes organizations that plan to give nothing, so a 3.5% budget does not guarantee that any individual gets 3.5%. Merit pay is meant to be uneven. Strong performers are supposed to receive more than the budget, and others less.

The second thing is that this is not a dramatic change. The 2027 figure sits only about a tenth of a point above what employers report actually paying in 2026, and it is close to what they paid in 2024 and 2025. The headline is not "raises are collapsing". It is "raises are staying stuck at a modest level".

Does a modest level mean flat real wages? That depends on prices. Take an illustration, not a forecast. If prices rise 3% over the year and your pay rises 3.5%, you are about half a percentage point ahead. On an $80,000 salary, that is a gross rise of $2,800, of which perhaps $400 is real purchasing power. If prices rise 4%, you went backwards. The honest summary is that a budget in this range roughly tracks inflation, it does not get you ahead of it, and each year it does so, the market moves a little further from you if your role is hot.

Why across-the-board raises are shrinking

The survey, as reported, found about a third of companies planning across-the-board increases, 32% for 2027 compared with 36% a year earlier. I am passing along the reported figures as I could find them, and I would treat the exact percentages as approximate.

The direction makes sense once you see the incentive. An across-the-board raise spreads a fixed pool evenly, which is simple and also expensive, because it pays the same to people the company most wants to keep and people it does not. A merit-only approach lets the same budget be concentrated. When money is tight, companies tend to prefer that.

For you, the implication is uncomfortable but useful. If the budget is divided by performance, the rating you get matters more than it did, and so does the story that supports it. A quiet year of good work, never written down, competes badly against a louder colleague's year.

Leverage that still works in a modest-budget year

A small pool does not make every conversation futile. It changes which conversations work.

A change in scope. If the job you do now is materially bigger than the job you were hired for, you are not asking for a raise, you are asking for the role to be priced correctly. Managers can often get a reclassification or promotion approved from a different budget than the annual merit pool.

Market evidence. A figure from outside the building carries more weight than a feeling from inside it. We will return to this below.

Retention risk, honestly stated. If you have a competing opening in front of you, that is leverage, but only if you are genuinely prepared to take it. Bluffing is a poor strategy and easily detected.

Rare or critical skills. Industry spread matters. The survey notes higher merit budgets in areas such as high tech and banking, and lower ones in retail and healthcare. Where your sector sits changes how much room your manager has.

One-time money. If base pay is truly capped, ask about a bonus, extra paid leave, a training budget, or a title. They are not equivalent to salary, because base compounds into every future raise, but they beat nothing.

When the math favors looking outside

Here is the arithmetic that I find clarifying. Suppose pay for your kind of role has been rising about 5% a year in the market, and your employer gives you 3.5%. After three years, the gap between you and the market is about 4.4%: (1.05 divided by 1.035) raised to the third power. It does not look like much in any single year, which is why it goes unnoticed.

Moving jobs resets that gap in one step, and people who change employers often see a larger jump than people who stay, though I do not have a figure I would quote with confidence, and the size varies a great deal by field and year. The tradeoffs are real as well: a new employer means a new manager, a fresh start with less goodwill, and sometimes a worse fit.

A reasonable rule of thumb is that if you have asked clearly, with evidence, at the right time, and been told no or been offered a few percent, the useful question changes. It is no longer "how do I persuade them?" It becomes "what is the cost to me of staying at this price?"

Timing the ask around the planning cycle

This is the part most people get wrong. Compensation budgets for the following year are typically proposed and approved in the second half of the year. In the Mercer survey, most organizations described their 2027 budgets as still preliminary as of July. That means the window for influencing the number is before it hardens, not at your review.

Late summer to early autumn: budgets are being drafted. Your manager is asking what their team needs. Raise your case here, informally.
Autumn: budgets are being finalized. This is the last cheap moment to ask. If you are reading this in late October, you are in it.
Winter, at the review: the pool is set. You are now competing for a slice. Expect less movement.

Practically, this means you should ask for a conversation about your compensation a few weeks before your manager has to submit their numbers, and you should make it easy for them to say yes by doing their homework.

A one-page raise-ask brief

Write the following as a single page and give it to your manager ahead of the meeting. Notice that nothing on it is about what you need. Personal need is real, but it is not persuasive to a budget committee. Market rate is.

The ask, in one line. A specific number or range, and an effective date. Example: "I'd like to discuss moving my base salary to $X, effective January 1."
My role and scope. Title, what I own, what I have taken on in the past twelve months that was not in the original description.
Results. Three to five outcomes with a number or a clear before-and-after. Skip adjectives.
Market data. Two or three independent sources for comparable roles in your location and sector. Government wage statistics, industry salary surveys, and posted salary ranges on current job listings, which are now required in a number of places, are all usable. State the range, the source, and the date.
Where I sit against that data. One sentence: "My current pay is about X% below the median for this role."
The ask restated, plus a fallback. What you would accept if the full number is not possible: a smaller increase now with a dated review, a title change, or non-salary items.

It should fit on one page, and your manager should be able to forward it, unedited, to the person who controls the budget. That is the test. If they cannot, you have written it for yourself.

I came to salary conversations late, as many engineers do, with the quiet belief that good work would be noticed and rewarded without anyone needing to say so. Sometimes it is. More often, someone else on your team had a written page.

Frequently asked questions

Is a 3.5% raise good?

It is average for a budget, and whether it is good depends on inflation, your market, and your performance. If prices are rising by about three percent, it keeps you roughly level with a small gain. If your role's market pay is rising faster, it quietly widens a gap.

Should I ask for a raise even in a tight year?

Yes, if you have evidence and the timing is right. A tight pool changes the tactics, not the legitimacy. Ask for a specific number backed by market data, and be ready to discuss alternatives.

What if my manager says there is no budget?

Ask what would have to be true to change that, and when budgets are next set. Ask for a dated review instead of a vague promise. Then consider whether a non-salary item or a reclassification is possible.

How do I find market-rate data?

Use more than one source: government wage statistics for your occupation, industry salary surveys, posted salary ranges on current job listings, and conversations with recruiters. If sources disagree, say so, and use the range.

When should I start looking for a new job instead?

When you have asked clearly with evidence and the answer is still below market, and you can see that the gap will keep growing. Look while you are employed, since that is when you have the most choice.

A budget is a number someone typed into a cell. Nobody typed in the work you did, which is why you will probably have to be the one to put it on the page.