The Real Cost of Switching Jobs vs Getting Promoted in Place

Somewhere in the last five years, "just switch companies" became the default career advice for software engineers who want to make more money. And for a while, the data backed it up. During the Great Resignation peak in 2022, job switchers earned median raises of 16%, while stayers got roughly 4%. The math was simple. Stay and you fall behind. Leave and you get paid.

That era is over. ADP Pay Insights data shows the switching premium has collapsed to just 1.9 percentage points as of early 2026. In some sectors, including IT, stayers actually outearned switchers. The gap between leaving and staying is the smallest it's been since November 2020.

But this article isn't about whether you should stay or go. It's about the costs that most engineers never calculate before making that decision.

The costs of switching that nobody talks about

The salary bump gets all the attention. Here's what doesn't.

Equity forfeiture

When you leave a company, your unvested RSUs vanish. Gone. If you're two years into a standard four-year vest at Google (which front-loads at roughly 33/33/22/12), you've collected the majority of your initial grant. If you're 18 months in, you're walking away from potentially hundreds of thousands in unvested shares.

Your new employer will give you a new grant with its own vesting schedule. But your equity clock resets to zero. You start over. And at most companies, the initial grant is significantly larger than the annual equity refresh you'd receive as an existing employee. So you trade a maturing equity position for a fresh one that won't fully vest for another four years.

The engineers who run this math before accepting an external offer often find the net comp increase is much smaller than the headline number suggested.

The ramp tax

Software engineers take three to nine months to reach full productivity at a new company. HackerNoon's analysis of onboarding data puts the typical ramp at about six months for complex systems. During that time, you're learning the codebase, figuring out the deployment pipeline, and building context on why things are the way they are.

You're not building a promotion case during those months. You're surviving. And at most companies, you won't be eligible for a performance-based equity refresh until you've completed at least one full review cycle. That's 12 to 18 months before the new compensation machine starts compounding in your favor.

Relationship capital resets to zero

At your current company, your manager has watched you deliver for two years. Your skip-level has heard your name in calibration. Your peers can vouch for your work in promotion committees. That political capital took years to build, and it's completely non-transferable.

At your new company, nobody knows you. Your manager has no data on your trajectory. Nobody in leadership has context on your capabilities. You're back to proving yourself from scratch, which means the first promotion at the new company typically takes longer than it would have at the old one.

The risk you can't interview away

You can run five rounds of interviews, talk to the hiring manager, stalk Blind threads about the team, and still end up on a team with a bad manager, a decaying codebase, or a company that's about to do layoffs. This risk is asymmetric. You know exactly what you have. You don't know what you're getting.

The costs of staying that people underestimate

The switching costs are real. But so are the costs of staying too long.

Comp stagnation is compounding against you

If you've been at the same level for two or more years, the gap between what you're earning and what you could be earning is growing every single cycle. Raises, equity refreshes, and bonuses at the higher level are all calculated on a higher base. The delta doesn't stay flat. It accelerates.

At Google, the median total comp difference between L4 and L5 is roughly $125K per year. That's not a one-time loss when you miss a cycle. It's $125K plus the compounding effect on every future raise, refresh, and bonus. Over three years, the cumulative gap exceeds $160K even in conservative estimates.

The "one more cycle" trap

On Team Blind, engineers describe staying for "one more promo cycle" for years, resetting the timer every time it runs out. Each time, the reasoning sounds rational. "My manager said next half." "The project I need just started." "Review season is only four months away."

Three years later, they're still at the same level, having spent years accumulating a cost they can calculate in retrospect but couldn't see in real time.

The perception penalty

Engineers who stay at the same level for an extended period develop a reputation in calibration. The narrative shifts from "rising contributor who's ready" to "solid performer who's plateaued." Whether that label is accurate doesn't matter. It affects how much political capital your manager is willing to spend on your case, and how seriously promotion committees take the nomination.

The comparison, side by side

Here's how the two paths stack up across the dimensions that actually matter.

Dimension	Stay and promote	Leave for higher level
Compensation timeline	Slower ramp. Internal promotion raises are typically 10-15% of base.	Faster initial bump. External offers can be 20-40% higher total comp.
Equity position	Existing grants continue vesting. Refresh targets increase at new level.	Unvested equity forfeited. New grant starts from zero with 4-year vest.
Ramp time	None. You're already productive, known, and delivering.	3-9 months before full productivity. 12-18 months before first refresh.
Relationship capital	Intact. Manager, skip-level, and peers have context.	Reset. Nobody knows you. Promotion advocacy starts from scratch.
Risk	Low. You know the environment, the team, and the culture.	Moderate to high. Bad fit, surprise layoffs, and bad managers are real possibilities.
Promotion timeline	Depends on how close you are. Could be next cycle or indefinite.	You may arrive at the higher level on day one, but the next promotion takes longer without context.
Career velocity	Slower if you're stuck. Faster if you're genuinely close.	Faster short-term. Potentially slower if you spend 12+ months ramping.
Best when	You have a credible path and a manager who advocates for you.	You've been stuck for 2+ years, the ceiling is structural, or external comp is significantly higher.

When the math says leave

Not every stay-or-leave decision is close. Some situations have clear answers.

You've been at the same level for 3+ years with no credible path. If you've been passed over multiple times despite strong reviews, or if the structural ceiling is real and not just a rough patch, the compounding cost of staying will exceed the switching cost within a year or two.

The external market is paying significantly more for your skills. Check Levels.fyi for current comp bands at your level and target level. If the delta between your current total comp and what you'd earn externally is large enough to absorb the equity forfeiture, ramp time, and risk, the numbers speak for themselves.

A specific company will hire you at the next level. This is the single strongest financial argument for switching. You skip the internal waiting period entirely. Instead of spending another 12 to 18 months building a case at your current company, you arrive at the new one already at the higher level, with a fresh equity grant sized for that level.

When the math says stay

You have significant unvested equity. If you're sitting on $150K to $300K in unvested RSUs that vest over the next 18 months, the external offer needs to clear that hurdle before you break even. Most engineers who run this math find the gap is smaller than they expected.

You're genuinely close to promotion. "Close" means your manager has explicitly said you're being nominated in the next cycle, not that you feel ready. If the promotion is one cycle away and your manager is actively sponsoring you, the ramp cost and relationship reset of switching will likely cost you more time than waiting.

Your current environment is strong. A good manager, a team that gives you scope, and a company where the next level actually exists. Those aren't things you should trade lightly. They're harder to find than a higher salary.

The real framework

The decision isn't "which path pays more." It's which path's costs you can absorb.

Cost of staying = (comp gap per year x years you'll remain stuck) + opportunity cost of slower career trajectory

Cost of leaving = forfeited equity + ramp months of reduced productivity + relationship capital rebuild time + risk of bad fit

When cost of staying exceeds cost of leaving, switch. When cost of leaving exceeds cost of staying, stay and build the case.

The break-even point shifts based on four variables: how long you've been stuck, how much unvested equity you'd forfeit, how realistic internal promotion is in the next cycle, and how much the external market is paying for your level.

Run the numbers on your specific situation. The answer is in the spreadsheet, not in a Blind thread.

---

CareerClimb helps you build the documented case that makes the stay-or-leave decision clearer. Log your wins, track your trajectory, and see whether the path forward at your current company is real or a dead end. Download CareerClimb