Job-Hopping vs. Staying: The Bigger Offer Isn't the Raise

A recruiter’s note lands, or you finally open your LinkedIn messages, and now there’s a number on the table that’s bigger than what you make today. The instinct - a very human one - is to treat that gap as the raise. It usually isn’t.

Last time, in When to Leave a Team You’ve Outgrown, Even If You Like Your Manager, I wrote about the moment you realize you’ve outgrown a team, even one with a manager you like, on a codebase you don’t hate, surrounded by people you’d still grab lunch with. That’s a real decision, and it’s mostly an emotional one: you’re weighing loyalty and comfort against a ceiling you can feel closing in. At some point, though, the emotional decision turns into a financial one.

So back to that gap. It’s a smaller raise than it looks like, and sometimes not a raise at all once you price in everything the headline number doesn’t show you. I’ve made this move twice in my career. I’ve turned down offers I should have taken, and taken one I probably shouldn’t have. The common thread every time: I ran the comparison at the level of “new base + new bonus + new grant” vs. “current base + current bonus + current grant,” and called it a day. That’s not the comparison. Here’s the one that actually matters.

The switching premium isn’t what it used to be

For most of the 2021-2022 hiring cycle, job-hopping was close to a free lunch. Pay-data trackers were showing job switchers pulling down wage growth in the high teens while people who stayed put were stuck in single digits. The spread was wide enough that “just leave” worked as blanket financial advice, whatever the emotional cost.

That gap has narrowed a lot. In January 2026, ADP Research Institute’s Pay Insights had job-stayers at 4.5% annualized pay growth versus 6.4% for job-changers: a switching premium of about 1.9 percentage points, the smallest gap ADP has recorded since it started tracking this in 2020. It has widened back out since, to 4.4% versus 7% in ADP’s July 2026 reading, but a premium of two or three points is a different world from the one the “just leave” advice was built in.

I’m not bringing this up to talk you out of moving. I’m bringing it up because the entire “just take the bigger number” heuristic was built during a period when the bigger number was reliably bigger by a wide enough margin to absorb everything else. That margin is thinner now, and in some labor markets and role types it’s close to gone, which means the stuff that heuristic let you ignore is now the stuff that decides the outcome.

What the headline offer number doesn’t show you

Every offer comparison starts the same way: total comp at the new place minus total comp at the current place. That’s the number recruiters quote, the number you screenshot for your partner, the number that feels like the whole story. It isn’t. Here’s what it’s missing.

Unvested comp you’re walking away from. Most compensation structures that include equity, deferred bonus, or long-vesting retention grants are built to punish an early exit. That’s the point of them. Whatever you have sitting unvested at your current employer, vesting on a schedule you’re partway through, is gone the day you leave, full stop, regardless of who’s making the counteroffer. This isn’t a company-specific quirk; it’s how deferred comp works everywhere. Forfeiture here is often a meaningful five-figure amount, not a rounding error, and exactly how much depends entirely on your grant size, vesting schedule, and how far into the cliff you are. So the only number worth using is yours: pull up your own vesting schedule and find the dollar figure you’d be forfeiting on your actual exit date.

The new-hire grant vs. the refresh-cycle mismatch. New employers tend to front-load your first grant to make the offer look competitive, then settle into a smaller, standard refresh cadence in year two and beyond. Meanwhile, your current employer’s refresh cycle, the one you’re about to walk away from, was presumably already trending upward if you’ve been performing well, because that’s what refresh cycles are designed to do. Compare the new offer’s year-one number against your current employer’s year-one number and you’re comparing a front-loaded number to a steady-state one. Compare four-year totals, not first-year totals.

Ramp-up risk. Your performance review at your current job is anchored to a track record. At a new company, you’re starting that clock over, on a codebase you don’t know, under a manager who’s never seen you work, during whatever probationary or first-review window that company runs. A weak first year turns into a bonus that lands small or not at all, and a first refresh that goes the same way. This is a real, underpriced tail risk in every job-hop comparison, and it’s the kind of thing a spreadsheet with a single “expected” column will hide from you. Model a bad first year explicitly, not just the best case the recruiter described.

Benefits and cost resets that don’t show up in a comp number at all. New health plan, new deductible you’re paying into from zero partway through the calendar year, PTO balance reset to zero, any employer retirement-match vesting clock restarting. None of it appears on the offer letter, and all of it is real money.

The job search itself has a cost. Time you didn’t spend deep in your current codebase building the track record that turns into your next promotion. Negotiating leverage you give up by accepting quickly instead of running a real process. If you’re moving for a lateral title at a meaningfully bigger number, that’s usually fine. But if the new offer is also a title bump, ask honestly whether you’re ready to perform at that level on day one under a new manager’s judgment, or whether you’re buying a harder ramp on top of everything else.

The other side of the ledger: staying has a cost too

None of this is an argument for reflexively staying, and the current pay data doesn’t actually support that reading uniformly either. Bank of America Institute’s 2026 “Should I stay or should I go?” analysis of internal deposit data found that, in aggregate, switchers were still outgrowing stayers in Q1 2026 (8% after-tax wage growth versus 5%), but among the top 5% of earners the pattern flips hard, with stayers seeing gains near 10% against less than 2% for switchers. That’s the one income tier where staying now clearly wins.

Compression is real for everyone else: unless your current employer runs disciplined, market-anchored compensation reviews, your pay drifts below what the same role would cost to hire externally, because raises are political and slow while market rates move continuously. That drift compounds every year you don’t check it against the outside market, which is part of why even people who intend to stay long-term should interview periodically. It’s a pricing exercise, not just a threat.

Building the actual comparison

The version that holds up is the one I wish I’d run instead of the two-line comparison I described at the top:

1. Annualize both offers over the same multi-year window - three or four years, matched to your current employer’s vesting cadence - instead of comparing single-year snapshots.
2. Subtract unvested comp you’d forfeit from the new offer’s total, as a one-time hit in year one.
3. Haircut the new offer’s out-year projections for ramp-up risk. Don’t assume the recruiter’s best case; assume a below-target first year is at least plausible and weight for it.
4. Add back what staying is actually worth: not just current pay, but a realistic internal trajectory if you asked for a market-rate adjustment or already have a review coming, so you’re not comparing a real new offer against an artificially frozen current baseline.
5. Only then compare the multi-year totals, not the headline numbers.

This is the same exercise as evaluating an offer’s shape rather than its headline number. If you haven’t already, What “Good” Comp Actually Looks Like is the natural companion piece to this one.

A rough heuristic, once you’ve done the math

If the multi-year, risk-adjusted gap is still wide (a clear step up, not a marginal one) after accounting for everything above, take it. That was true even when the premium hit its record low. Aggregate averages compressing doesn’t mean every individual offer compressed with it, and a genuinely strong offer for the right match is still a strong offer.

If the gap is narrow once you’ve subtracted unvested comp and haircut for ramp-up risk, the deciding factor should usually be the one you started with, before the numbers got involved: the actual day-to-day, and the ceiling. Marginal dollars in a narrow-gap scenario are close enough to a coin flip that they shouldn’t be the tiebreaker.

Everything above is doable in a spreadsheet, and the first time an offer lands, you should build one from scratch, with your actual vesting schedule and your actual numbers, not rounded guesses. The problem is the second time. A recruiter reaches out again in eighteen months, there’s a decision date on the offer, and you’re hunting for that old spreadsheet, trying to remember what’s vested since, eyeballing your comp history instead of pulling it up. That’s the shortcut this whole post has been arguing against, and it’s an easy one to fall back into when the real numbers are scattered across old offer letters, a vesting portal you log into once a year, and a retirement account you check even less often than that.

The fix is boring, and it only has to happen once: set up a net-worth and compensation tracker while nothing is on the line, so your vesting schedule, comp history, and account balances are already sitting there and already current the next time a number lands in your inbox. That turns the five-step comparison above from a weekend project into a five-minute update, which is the difference between actually running the multi-year math and deciding on the headline number because the deadline is Friday.

The takeaway

A bigger number is not automatically a bigger raise. It’s the start of a comparison, not the end of one. And the difference between “looks like a raise” and “is a raise” lives in exactly the stuff most people skip because it’s harder to look up than a base salary line. Do the multi-year version. Most of the time it still points the same direction as your gut. Occasionally it doesn’t, and that’s the one time this exercise actually pays for itself.