Executive search

One role, six months, and one submission you do not get back. one submission you do not get back

1

role open, and no second req to hide behind

The JD cannot go on a board, so nothing arrives on its own. The person you want is employed, senior, and reads your first line between two meetings.

Start a confidential search
Executive search on Recruitkar

Two hundred and forty names to earn one

Say the market for this title in India runs to about two hundred and forty people. The client asked for the map, not a Friday shortlist, so you have to look at every one of them before you can honestly call the shortlist a shortlist. Sourcing is 1 credit per candidate, refunded when the person cannot be found and refunded again when they come back under the quality threshold, so the map costs roughly what the arithmetic says it costs.

Each mark is one name on the market map. The solid ones are the share anyone actually gets to.

240

names you have to look at before you trust the map

0

that arrive on their own, because the role cannot be posted

1

profile you can afford to put in front of the client

Three things that give way in a search this long

None of this breaks on a 40-position volume drive. It breaks here, because one role runs for months and every name has to stay in your head the whole time.

around name 60

The map stops fitting in your head

By the sixtieth name you have opened the same profile twice and not noticed. One tab came off LinkedIn in April, one off an old Naukri import, one out of the CV folder the client's HR sent before they went quiet. You start a second sheet to keep track of the first sheet. Recruitkar folds all six intake routes into one ranked list, so a person who arrives three ways is one card carrying all three sources.

week three, when your own archive runs dry

Nothing is coming in

The search is confidential, so the JD goes nowhere. No board, no walk-in, no apply link, no referral note from the client's own team. Your archive carries the first fortnight, then the folder simply ends, and the pipeline is only what you go out and take. Discovery across external profile databases, LinkedIn import, extension imports from Naukri, Indeed and Shine, and a plain resume-folder upload all land on the same board, so outbound stays the only route and still adds up to a map.

the second submission

Your own confidence in the shortlist

The first profile you send is read on merit. If it misses, the second one is read for reasons to say no, and you feel that shift on the call. So you sit on a good profile for two extra days because you cannot afford to be wrong twice. Every candidate is scored against the actual JD across about ten dimensions, with matched and missing skills named one at a time, red flags and a verdict from strong fit to not fit, and a member's candidate can be held until a teammate approves it.

How the six months actually run

The client wants a market map first and a name much later. The order below is built for that, not for a Friday shortlist.

day 1-4

Build the map before you write to anyone

Nobody gets a message this week. You empty your archive into the job first: resume-folder upload at 0.5 a screen, board imports at 0.5, LinkedIn imports at 1, then discovery across external profile databases at 1 a candidate, refunded when the person cannot be found or comes back under the quality threshold. The apply link stays switched off, because there is no version of this search that is public.

day 5-12

Score it and tier it

Every name gets scored against the real JD, not a title string, across roughly ten dimensions. What comes back is the map the client asked for, in tiers, with a reason under each name and the missing skills spelled out. That is the document you present in week two, and it is the reason the next four months get bought.

week 3-10

Approach, and keep approaching

Contact lookup is 1 credit and only charged when a contact is actually found. Email is 1 credit per candidate ever, not per message, so the third follow-up in August costs nothing more than the first one in June. WhatsApp templates at 1 credit for the ones who read on their phone, a 24-hour session when they reply, and AI voice calls at 2 credits a connected minute for the people who will never open a mail.

week 6 to DOJ

Submit, then hold it together

A teammate signs off internally before anything goes out. The client sees only the fields you ticked, and their approve, reject or note lands straight back on the candidate. Then the slow part: 90 days notice at this level, a buyout conversation, a counter-offer in week eleven. The seven-stage board holds the position while the calendar drags.

The client wants a market map first and a name much later. The order below is built for that, not for a Friday shortlist.

What holds when the search runs long

One person stays one card, five months later

The same VP can reach you off LinkedIn in March and again through a Naukri import in July. They do not become two rows and a duplicate mail. They stay one card with one history, one score against this JD, and every stage already passed.

Persistence is not billed twice

Email is 1 credit per candidate for the life of the job, however many times you write. Contact lookup only charges when a contact is found, and sourcing refunds when the person is not. If the search dies and the client reopens it as a different req, reusing that candidate on another of your own jobs is free and re-scores them against the new JD.

Nothing reaches the client unchecked

Internal review can hold a member's candidate until a teammate approves it, so a junior's enthusiasm does not turn into your credibility problem. Client sharing shows only the fields you chose, and per-member credit caps mean nobody burns the search budget on a hunch.

The four things people ask before they start

The client will not let the role be named anywhere. Can I run this at all?

Yes, because nothing here needs to be public. The JD lives inside the job, the shareable apply link simply stays off, and the whole pipeline is built outbound. When you share a candidate, you tick which fields the hiring manager can see and they get a page with only those, plus approve, reject and a note box. What Recruitkar cannot control is what a candidate works out from a conversation with you.

I am writing to a country head. If it smells like a mass mail, they are gone.

Then do not send them one. Being straight about it: our email is a blast tool, with merge fields, a deliverability pool and bounce monitoring. It is not per-candidate AI writing, and no domain warming is bundled. For your top twenty, write each mail yourself and use the scorecard for material, since it names their matched and missing skills and suggests questions for that person. Keep the blast for tier three, where the alternative is not writing to them at all.

The client gave me four lines on a call and refuses to put a JD in writing.

Then the scores stay honest about it. A two-line JD does not get a confident verdict, and those dimensions come back empty instead of inflated. The flip side is real: vague JD in, vague verdicts out. The useful part is that the gaps show on screen, which is a better thing to take back to the client than a shortlist they will reject for reasons they never wrote down.

Am I supposed to put a CFO candidate through an AI interview?

No, and we would not suggest it. AI voice and video interviews, in 11+ languages with recording, proctoring events, a scorecard and a recommendation, cost 4 credits a minute and earn their keep on a 200-CV role. A candidate at this level will read one as an insult. Use human scheduling instead, off Google Calendar or Outlook, on Meet, Teams, Zoom or in-office, and save the AI interviews for the volume req the same client sends you next quarter.

Bring one role and the four lines you have.

Start with the map. The map is what buys you the next four months, and it is the part that takes the longest by hand.

Get startedTalk to us first