Screening for Budget Fit Before You Waste the Shortlist

Screen for budget fit in the first ninety seconds of contact, not after round three. Ask for expected CTC as a range, ask what that range means in fixed versus variable, and ask the notice period in the same breath. If the expectation sits above your band, say the band out loud immediately. A candidate who walks away in minute two costs you nothing. One who walks away after the panel round costs you three weeks.

Every recruiter has done this. You run a clean screen, you shortlist five genuinely strong people, you send them to the hiring manager — and then compensation comes up and all five are forty percent above the band. The screen wasn't bad. The screen was wasted, because it optimised for the wrong variable in the wrong order.

The gate is in the wrong place

Most Indian hiring funnels put the money conversation just before the offer, because that's where it feels polite. That's exactly backwards. Skill match is expensive to evaluate and cheap to be wrong about. Budget fit is cheap to evaluate and catastrophic to be wrong about.

Moving one gate is the entire intervention. It is not sophisticated. Most teams still don't do it, because asking about money early feels like leading with the least flattering thing about the job.

The question that isn't crass

The crass version is "what's your current CTC?" — asked flatly, first, with no reciprocity. It reads as an attempt to anchor the offer to what the candidate is already being underpaid, which is often exactly what it is.

The version that works — and that good candidates respect — has three properties: you go first, you ask about expectation rather than history, and you ask for structure rather than a single number.

"Before we go further, let me be transparent about the band so neither of us wastes time. This role is budgeted at ₹X–₹Y fixed, plus a variable component. Does that range work for what you're looking for? And what's your notice period?"

Three things happened there. You disclosed first, which converts an interrogation into an exchange. You named fixed, not CTC, which is the only number that means anything (more on that below). And you bundled notice period in, so the money question doesn't sit alone as the whole point of the call.

If the candidate deflects — "I'd rather understand the role first" — that's legitimate, and you should let them. But hold the line on the gate: "Completely fair. I'll only ask that if my range is a non-starter for you, tell me now rather than in week three." That framing gives them an honest exit and almost always gets an honest answer.

Ask for expected, not current. India has no salary-history ban of the kind several US states have legislated, so asking current CTC is legal here. It's still a bad screen. Current CTC tells you what someone's last employer got away with paying. Expected CTC tells you whether they will accept your offer. Only the second one is decision-relevant, and asking only the second one keeps you clear of the pay-equity problem discussed at the end of this post.

"Eighteen lakhs" is not a number

The reason a budget screen so often fails even when you did ask is that CTC is not a price. It's a sum of a fixed component, a variable component, statutory employer contributions, and a pile of notional benefits — and two companies can quote the same CTC while the money that actually reaches the candidate's bank account differs by lakhs.

<text x="140" y="72" text-anchor="end" font-weight="600">Company A</text>
<text x="140" y="88" text-anchor="end" fill-opacity="0.65">product firm</text>
<g stroke="currentColor" stroke-opacity="0.55" fill="currentColor">
  <rect x="150" y="60" width="389" height="40" fill-opacity="0.55"/>
  <rect x="539" y="60" width="56"  height="40" fill-opacity="0.32"/>
  <rect x="595" y="60" width="25"  height="40" fill-opacity="0.18"/>
  <rect x="620" y="60" width="10"  height="40" fill-opacity="0.10"/>
  <rect x="630" y="60" width="21"  height="40" fill-opacity="0.04"/>
</g>
<text x="663" y="78" font-weight="600">₹1.17L</text>
<text x="663" y="93" fill-opacity="0.65">fixed / month</text>

<text x="140" y="162" text-anchor="end" font-weight="600">Company B</text>
<text x="140" y="178" text-anchor="end" fill-opacity="0.65">services firm</text>
<g stroke="currentColor" stroke-opacity="0.55" fill="currentColor">
  <rect x="150" y="150" width="319" height="40" fill-opacity="0.55"/>
  <rect x="469" y="150" width="97"  height="40" fill-opacity="0.32"/>
  <rect x="566" y="150" width="18"  height="40" fill-opacity="0.18"/>
  <rect x="584" y="150" width="7"   height="40" fill-opacity="0.10"/>
  <rect x="591" y="150" width="59"  height="40" fill-opacity="0.04"/>
</g>
<text x="663" y="168" font-weight="600">₹0.96L</text>
<text x="663" y="183" fill-opacity="0.65">fixed / month</text>

<g stroke="currentColor" stroke-opacity="0.55" fill="currentColor">
  <rect x="150" y="222" width="14" height="14" fill-opacity="0.55"/>
  <rect x="256" y="222" width="14" height="14" fill-opacity="0.32"/>
  <rect x="376" y="222" width="14" height="14" fill-opacity="0.18"/>
  <rect x="452" y="222" width="14" height="14" fill-opacity="0.10"/>
  <rect x="546" y="222" width="14" height="14" fill-opacity="0.04"/>
</g>
<g fill="currentColor" font-size="12">
  <text x="170" y="234">Fixed pay</text>
  <text x="276" y="234">Variable / bonus</text>
  <text x="396" y="234">Employer PF</text>
  <text x="472" y="234">Gratuity</text>
  <text x="566" y="234">Insurance, perks</text>
</g>
<text x="150" y="268" fill-opacity="0.7">A candidate quoting "18 LPA" from Company B is asking you for ₹0.96L/month.</text>
<text x="150" y="286" fill-opacity="0.7">A candidate quoting "18 LPA" from Company A is asking for ₹1.17L. Same words, ~₹2.5L/yr apart.</text>
<text x="150" y="310" font-size="11" fill-opacity="0.55">Illustrative package structures for arithmetic, not survey data.</text>

Two of those components are statutory and worth knowing precisely, because they appear on every CTC sheet in the country and candidates rarely count them as pay:

• Employer PF — 12% of basic wages, mandated by the Employees' Provident Funds and Miscellaneous Provisions Act, 1952. Real money, but it isn't in-hand this month.
• Gratuity accrual — the roughly 4.81%-of-basic line you see on CTC sheets is derived from the Payment of Gratuity Act, 1972 formula of 15 days' wages per completed year. It's real, but it vests only after five years of continuous service. For a candidate who plans to switch in three, it is worth exactly zero.

So the screening question is not "what's your expected CTC." It's "what fixed do you need?" That single reframe removes most of the ambiguity that causes late-stage blowups.

Why the same role has a different band in a different city

An engineer with the same skills, same years, and same output does not command the same number in Bengaluru as in Indore. That's not a moral fact about the engineer; it's a fact about the local price of talent, driven by how many other employers within commuting distance are bidding for the same person.

Practically, this means two things for your screen:

1. A candidate's current CTC is a function of their current city, not their current ability. Someone moving from a Tier-2 city into a metro role is right to ask for a large jump, and treating that jump as a red flag will make you reject good people. Someone moving the other way may accept less than your band and still be delighted — and you should not exploit that (see the ethics section).
2. "Remote" does not delete the city variable, it just makes you choose one. If you're hiring remote, decide before you screen whether you pay a single national band or a location-adjusted one, and say which. Discovering that you never decided, while a candidate in Pune and a candidate in Gurugram are both in your final two, is a self-inflicted wound.

Notice period is a price variable, not an admin detail

In India this is the most under-used lever in the whole screen. A 90-day notice period is not just an inconvenience — it's a real cost to you (a role sits empty for a quarter) and a real constraint on the candidate (they lose negotiating flexibility the longer they're in limbo).

Treat it as part of the price:

• A candidate at ₹X with 15 days' notice and a candidate at ₹X+10% with 90 days' notice are not obviously ranked. Price the empty seat. If the role is blocking a delivery, the faster joiner may be worth the premium.
• Buyout is negotiable and often cheap. Many candidates can shorten notice by paying back their employer, and the amount is frequently smaller than a month of the hiring delay costs you. Ask: "If we needed you in 30 days, what would that take?" You will get a specific, actionable answer surprisingly often.
• A long notice period raises your drop-off risk. Ninety days is ninety days for the counter-offer to arrive. Weight it accordingly in the shortlist, not in the offer stage.

Weighting CTC expectation against skill match

The failure mode on both sides is treating budget fit as a binary. "Over band → reject" throws away the exceptional candidate who is worth an exception. "Great candidate → push it through" is how you end up with a compressed band and a retention problem twelve months later.

A workable model: score skill match and budget fit separately, then act on the pair.

Two things make this table usable rather than decorative.

First, you need a real definition of "must-have," which means someone must extract it from the job description before the screen starts — not vibes, an actual list. If you don't have one, run the JD through the keyword scanner and argue about the output; disagreeing with a machine's list is a much faster way to reach a real list than starting from a blank page.

Second, the exception has to be pre-approved, not retroactively begged for. The whole point of an early budget gate is to stop the escalation happening after emotional investment. Ask your hiring manager one question before you source: "What's the number at which you'd say yes anyway, and what would the candidate have to be?" Get the answer before you meet anyone. If the answer is "there is no such number," that's excellent — your gate is now hard and your screen is now fast.

The ethics, stated plainly

Budget screening goes wrong in one specific way, and it's worth naming.

Anchoring an offer to a candidate's current CTC launders existing pay inequity into your organisation. If a person was underpaid at their last job — for reasons that may include gender, background, a bad first negotiation, or simply a stingy employer — and you offer them "current + 15%", you have imported that underpayment and added your own margin to it. Do that across a team and you have built a pay structure whose logic is other companies' mistakes.

This isn't only a values question. The Code on Wages, 2019 prohibits discrimination on the ground of gender in wages for the same or similar work. A pay structure derived from candidates' prior salaries rather than from the role's value is a structure that can drift into exactly that exposure without anyone intending it.

The fix is the same practice that makes the screen efficient, which is a rare and pleasant alignment:

• Set the band from the role, before you see a single CV. Then the band is a fact about the job, not a negotiation about the person.
• Disclose it. If you can't say the range out loud, the range is probably not defensible.
• Ask expected, not current. You lose an anchoring advantage. You gain a pipeline that doesn't collapse at the offer stage, and a comp structure you can explain in an audit.
• Don't underpay the candidate who would have accepted less. You will pay for it in the exit interview, and you will pay more to backfill.

Note also that the candidate playbook is public and universally read. Indeed India's own candidate-facing guidance tells switchers they "can also add about 15% to 20% to your current CTC and mention a range." The person on the other end of your screening call has almost certainly read some version of that. Screening as if they haven't is not shrewd; it's just slow.

What to change this week

1. Add two fields to the top of your screening script: expected fixed (a range) and notice period (with buyout feasibility).
2. Get a pre-approved exception threshold from the hiring manager before sourcing. One question, one Slack message.
3. Rank the shortlist on skill match and budget fit as two columns, not one blended gut score. The pair is what the hiring manager actually needs to decide.
4. Stop asking for current CTC. It buys you an anchor and costs you a defensible pay structure.

None of this requires software. It requires the gate to move.

FAQ

Is it legal in India to ask a candidate for their current CTC? Yes. India has not enacted a salary-history ban of the kind several US states have. But legality isn't the argument — anchoring your offer to a candidate's prior pay imports whatever inequity produced that pay, and the Code on Wages, 2019 prohibits gender-based wage discrimination for the same or similar work. Asking for expected CTC gets you the only number you actually need to decide.

Won't disclosing the band mean every candidate just asks for the top of it? Some will, and that's fine — the top of your band is by definition a number you were willing to pay. What disclosure actually buys you is the candidates who don't ask, because they've self-selected out at minute two instead of week three. That saved time is the return.

How should I compare a candidate asking more but with a shorter notice period? Convert the notice period into money. Estimate what a quarter of an unfilled seat costs the team — delayed delivery, overtime on the rest of the team, a re-run of the whole search if they drop out during a 90-day wait. Compare that to the difference in ask. Often the faster joiner is cheaper even at a higher CTC. Also ask directly what a 30-day exit would take; buyouts are common and frequently smaller than the delay.

Can any of this be automated? Partly. Extracting the real must-haves from a JD and matching them against a stack of resumes is mechanical work that software does faster than a person — that's what ShortlistAI does, and it scores budget fit alongside skill match rather than after it, so an over-band candidate never reaches the top of your list by accident. What software cannot do is set the band, or decide when a candidate is worth the exception. Those are your calls, and they should be made before you open the first CV.