What a job switch should actually pay you in 2026

foundit ran its Appraisal Survey 2026 on more than 2,500 professionals in June, and Business Today reported the results on 23 July 2026: 70 percent of people expect at least a 21 percent salary hike when switching jobs, and one in five wants more than 40. Only 7 percent would take less than 10. Somewhere between those two numbers sits a real answer, and it depends entirely on which band you’re in.

Most people ask the wrong question here. They ask “what’s a good hike percentage” when the number that matters is your absolute CTC against the market rate for the role you’re moving into. A 40 percent hike on a bad base is still a bad offer.

Why percentage on CTC is the wrong unit

A 30 percent hike on 6 lakh gets you to 7.8. A 20 percent hike on 18 lakh gets you to 21.6. The second person did worse in percentage terms and is still earning three times more. Recruiters like quoting hike percentage because it sounds generous and costs them nothing to promise; candidates like it because it’s simple math. Neither reason makes it the right thing to optimize.

What you should actually compute is: new fixed CTC, minus current fixed CTC, divided by twelve, compared against your monthly expenses plus whatever you’re trying to save. Run the actual number through something like our CTC to in-hand calculator before you accept anything, because a 40 lakh offer that’s 35 percent variable pays out very differently than 34 lakh that’s mostly fixed.

The hike percentage is a negotiating anchor. It is not the thing you’re actually buying.

What’s a realistic salary hike when switching jobs, by band

These bands are rough, and India’s market is not one market, it’s Bangalore-product, Bangalore-services, Pune, Hyderabad, and everywhere else, each with its own gravity. Still, four patterns hold up consistently in 2026.

Fresher, 0 to 1 years. There’s usually nothing to negotiate a hike against because there’s no prior CTC worth comparing. The number you’re chasing is a market band for the role and company tier, not a percentage over college stipend or internship pay.

2 to 4 years. This band gets the biggest percentage jumps because the market has decided you’ve proven you can ship, but you haven’t specialized enough to be irreplaceable. 25 to 40 percent within services is normal. Moving services to product can go well past that, sometimes 60 to 80, because the base you’re leaving is genuinely low relative to product-company pay for the same skill.

5 to 8 years. Hikes compress here. Companies expect you to already be paid closer to market, so 20 to 30 percent is the honest range unless you’re changing something structural, like moving into a scarce specialization or taking on management scope for the first time.

8+ years. Percentage stops being a useful frame entirely. At this level you’re negotiating total comp packages with RSUs, bonus structures and sometimes a level bump, and a senior engineer moving from an IC track to an EM track can see comp move in ways that have nothing to do with a clean percentage over last CTC.

Services versus product is the bigger lever than experience

The single biggest determinant of hike size in India isn’t years of experience, it’s which side of the services/product line you’re standing on when you jump. Services companies pay a lower base for the same skill because their business model bills clients by headcount and margin, not by individual output. Product companies pay closer to what the work is actually worth to them.

That gap is why a services-to-product move produces the 50, 60, sometimes 100 percent hikes that make people ask if the number is a typo. It usually isn’t. It’s just that the starting point was underpriced, not that the new offer is generous.

Product-to-product moves, by contrast, tend to look a lot more like the 20 to 35 percent range, because both sides are already pricing you closer to the same market.

What actually justifies more than 50 percent

Three things reliably get you past 50, and none of them is “I asked nicely.”

A competing offer, in writing, with a number. Not a verbal mention that you’re “talking to a few places.” An actual offer letter changes the conversation from persuasion to matching, and companies match faster than they negotiate from scratch.

A title or scope change. Moving from senior engineer to staff, or from IC to a team-lead role with reports, changes the job you’re being paid for. The hike reflects the new job, not a reward for the old one.

A skill the market is currently short on. This one is real but it decays fast. LLM infrastructure and applied ML engineers commanded 50 to 80 percent hikes through parts of 2025 and into 2026 precisely because supply hadn’t caught up. Skills like that don’t stay scarce forever, and chasing the current hot skill purely for the hike is a bet, not a strategy.

Without one of these three, a request for 50-plus percent tends to read as either naive about the market or a bluff, and experienced hiring managers have seen enough bluffs to tell.

Anchoring your number to something real

The BLS puts the US median annual wage for software developers at $135,980 as of May 2025, growing 10 percent through 2035. That’s not directly usable for an India number, obviously, but it’s a useful sanity check on how fast the underlying demand for the skill is growing globally, which eventually shows up in what Indian product companies are willing to pay to compete for the same talent pool.

For an India-specific number, use three sources and see where two of them agree: a comp aggregator, someone who actually does the job at the target company right now, and the posted band if the company discloses one. AmbitionBox and Glassdoor are the obvious starting points, though take the self-reported numbers there with real skepticism: the sample skews toward people happy enough with their offer to bother posting it. If your target number is comfortably inside that overlap, it’s defensible. If it’s above all three, you’d better have one of the three justifications above.

Once you have two offers or an offer plus your current package, put them side by side properly rather than eyeballing the CTC line. Our offer comparator breaks down fixed, variable, and joining bonus so you’re not comparing a number that includes a signing bonus against one that doesn’t.

One thing I’d push back on

I think the advice to “always negotiate, never accept the first number” gets applied too uniformly. If a services company opens at 35 percent and your target band, from three independent sources, tops out around there, pushing for 45 just because a LinkedIn post told you to negotiate everything mostly signals that you didn’t do the homework. Negotiate hard when the gap between offer and market is real. When it isn’t, spend that energy on the interview loop instead, since that’s what got you the number in the first place. Our tech salary negotiation guide covers the actual conversation once you know what to ask for, and CTC vs in-hand salary covers what happens to the number after tax takes its share.

I don’t have good data on how often candidates who push past a fair anchor lose the offer entirely versus just annoy the recruiter. My guess is it’s more common than the negotiation-blog genre admits. If you’ve seen numbers on that, I’d want to see them.

What to do the week before you say a number

Pull three data points before the recruiter call, not during it. First, check what the role pays at the specific company, not the industry average, because a hot startup and a slow-moving MNC can differ by 40 percent for the identical title. Second, ask one person who already does the job there, even a weak-tie LinkedIn connection, what the honest range looks like. Third, decide your floor: the number below which you’d rather stay put than move.

Freshers skip step two most of the time, since there’s no one to ask yet, and that’s fine. Everyone past two years of experience has no excuse. A five-minute call with someone one level ahead of you at the target company beats an hour of scrolling comp forums, because forum numbers are unverified and often self-reported by people bragging.

Write your number down before the call. Not a range you’ll improvise around, an actual number, because people who wing it under mild social pressure from a friendly recruiter tend to undershoot their own research by 5 to 10 percent without noticing.

One more thing worth checking, and almost nobody does: your own last three appraisal letters. If your annual increments have consistently landed near the industry’s flat 9 percent, your employer already has a working model of how little pushback you give. That history doesn’t change what the new company should pay you, but it might explain why your current one never offered more without you asking first. Bring the letters. Numbers travel better than memory.