3 Reasons You Didn't Get the Raise You Asked For
Companies handle raises in wildly different ways. Some run a formal cycle every spring with calibration meetings and rating scales. Others manage it on an ad hoc basis right before a budget cycle. Some only hand out merit increases when someone gets an offer somewhere else.
About the only thing that stays consistent is the pool of money allocated for raises. Merit budgets in the US hover around 3% of payroll, and that amount is divided among everyone on the team. Some may get 5%, some may get 0%.
When your request comes back as "not right now" or "let's revisit next cycle," it doesn't necessarily mean you didn't deserve the raise. It might be that your timing was off or that your manager was not fully aware of your impact.
So, how do you ensure you communicate the right information at the right time to have the best chance of getting the raise you are looking for? Let's walk through 3 ways software engineers miss the mark when asking for more money.
1. You Didn't Track Your Achievements
Think back to what you worked on last March. Most of us can name the big project and maybe an incident we helped resolve. The rest fades.
That's a problem when your review lands in November, and you're rebuilding ten months of work from memory. What you hand your manager is a summary of what you happen to remember, which, at best, is incomplete and, at worst, incorrect.
There's also research showing that fuzzy ownership costs some people more than others. Heilman and Haynes ran three experiments where participants evaluated a man and a woman who had worked together on the same task. When it wasn't clear who contributed what, the woman was rated as less competent and less influential than her partner. When the individual contributions were well documented, that gap disappeared.
You want the work you have done attached to your name somewhere other than in your own notes, so you can be sure you get credit and avoid biases. Here are a few places you should be making sure you keep track of and are credited in:
- Design docs and RFCs. The author field shows you understand the problem and are actively proposing solutions.
- Postmortems. These list what you found, what you changed, and the impact. Great for showcasing the ROI of your work and attention to detail.
- Decision threads in Slack or on the PR. When you made the call on an approach, that thread is your evidence.
- Demo recordings. Five minutes of you walking through the work is much harder to reassign than a line item in a spreadsheet.
The great thing is that you probably don't have to create anything new here. You're putting your name on work you're already doing and making sure you can find it again in ten months.
Building a Year of Wins covers in more detail what should go in that log you keep.
2. You Focused on Effort Instead of Results
Your manager is splitting that 3% across a team where (hopefully) everyone worked hard. Everyone has a story about a difficult project or a weekend they gave up.
Effort is the baseline. It doesn't help your manager decide who gets 5% and who gets 1%.
What helps is being able to point to what impact your work made to the team, company, and customers:
- Deploys that took an afternoon now take twenty minutes.
- Support tickets for the billing flow dropped from 40 a week to 12.
- The Postgres upgrade completed without a rollback and with no customer-facing downtime.
The tricky part is work that doesn't produce a number. Reliability, security, migrations, and cleanup often have nothing measurable attached, and plenty of engineers can't see revenue or churn data at all.
It's important that you don't invent a number to fill that gap. If your manager sees a figure you can't back up, it gets challenged, and the rest of your case looks shaky by association.
When there's no metric, answer a different question. What would be worse today if you hadn't done the work?
- Before you implemented the logging updates, on-call spent the first 30 minutes of every outage figuring out which service was failing. Now they can jump right in where the error is occurring.
- Before the dependency upgrade, the team couldn't start the SSO work. That whole project was blocked until you identified and implemented an upgrade path.
- Before you documented the release process, one person on the team felt comfortable deploying to production. He took two weeks off in July, and nothing shipped. Now we have an entire team who shares the deployment process.
Those are results, and they hold up when your manager is considering your request for a raise.
Your Metrics Only Tell Half the Story goes deeper into how to connect your technical work to the outcomes leadership tracks.
|
|
Where can AI save you time?
My friends at Big Creek Growth put together a quick survey to spot the repetitive work you can hand off to automation.
|
|
3. You Asked Too Late
Compensation planning runs ~8 months ahead of the money showing up in your paycheck. Employers were already building their 2026 raise budgets in the summer of 2025, and those are usually locked in by the end of Q3.
So by the time you sit down for your review, the size of the pool was likely set months earlier. Your manager isn't deciding how much money exists. They're dividing up a number someone handed them.
That doesn't mean the door is closed. Budgets are revisited, and off-cycle adjustments occur throughout the year. But if your review is the first time salary has come up, your manager is hearing it when they have the least room to work with.
Two questions I always make sure employees ask and confirm every year:
When do compensation decisions get made, and what do you look at?
What would make this raise request an easy case for you to make?
Neither one is a big ask, and both work no matter how long you have been on the team. The answers tend to land in one of a few places:
- A specific date or quarter. Now you know the real deadline for the first two items on this list.
- A scope gap. You're being paid correctly for the work you're doing, and the work has to change before the number does.
- A peer comparison you didn't know about. Uncomfortable to hear, but much better in March than in November.
- "I'm not sure." Still useful. Your manager may not run this process, which tells you the conversation needs to happen earlier, with someone else, or both.
Doing all three of these improves your odds, but it doesn't guarantee anything.
Bankrate's 2025 pay raise survey found that 43% of US workers received no pay increase in the previous 12 months, up from 39% the year before. In a year like that, even a well-prepared case can still come back as a no.
The goal is to remove as many barriers as you can between you and that raise. Plenty of what goes into these decisions is out of your hands. Preparing your case and bringing it up at the right time are not. A few hours spread across the year can be worth thousands.