Most executives write a resume as if it were an application. At the retained level it is a dossier input, and that one difference changes how the document should be built, what it needs to say first, and what it can leave out.
The search starts with a map
A retained firm does not wait for resumes. It builds a long list by market-mapping. Researchers identify who holds the equivalent job today, who was recently promoted, who looks blocked, who spoke at the right conference, who sits on which board. Direct, confidential outreach to people who were not looking is the main route to a hire.
The long list runs 50 to 100 names. The short list is three to five, each with a detailed dossier. Vendor benchmarks put a director-level search at six to twelve weeks, a VP search at eight to fourteen, and a C-suite search at twelve to sixteen or more. Demand is not falling: Korn Ferry's executive search segment grew 9% in the fiscal year ended April 2026, with North America its fastest-growing region in the final quarter.
Your resume enters that process after you are already on the list. It is read first by a researcher or associate assembling your dossier, then by the partner preparing the client presentation, then by the client. It is a verification and packaging document. The discovery document is your LinkedIn profile, which is why the two must say the same thing, and why the profile gets rebuilt once the resume is final.
What that means in practice. The document is being read by someone who is building a case for you and needs the evidence laid out so they can lift it. Write it as the evidence file for that case.
Reader one: the researcher
The researcher's job is to fill a dossier template. Title, company, dates, scope, reporting line, the size of what you ran, the situations you have operated in. They are reading your resume for parseable facts and they are reading it fast.
This is where layout matters, and it is the only part of the six-second myth worth keeping. The one eye-tracking study anyone cites is from 2018 and measured a volume recruiter's triage pass, a different read from a partner reviewing a shortlisted executive. What survived it is about structure: simple layout, clear section headings, bold titles, bulleted accomplishments, and white space held attention; cluttered multi-column designs and long sentences lost it.
So the left-hand column of page one does a specific job. It carries proven scope in four numbers; the environments you have operated in (public, private, sponsor-owned, founder-led, turnaround, scale-up); six to eight core skills; and boards and advisory roles. A researcher can fill most of a dossier from that column alone, and if they cannot, they fill it from LinkedIn or guess.
Every role needs a scope line: reported to, P&L or budget, direct and indirect reports, geographic or functional reach. If you did not own a P&L, say what you did own. An omitted scope line reads as hiding something, and the researcher writes "scope unclear" in the dossier.
The database question, answered properly
You will be told that 75% of resumes are rejected by software before a human sees them. There is no primary source for that number. The best study available, a small 2025 panel of recruiters across ten platforms, found 92% of their systems do not auto-reject on formatting, keywords, or AI match scores. Where automated rejection exists it is for legal and compliance criteria: work authorization, licences, location.
At a search firm the applicant tracking system is a database of record: Invenias, Clockwork, Thrive, or Bullhorn at the firm, and Workday or Greenhouse at a corporate. A human searches it by keyword and Boolean. Two things follow. Your titles, dates, and employers must parse cleanly as text, with no headers in graphics and no tables that scramble. And the words a researcher would search for must be present in the words you use. At this level, parsing is what matters.
Reader two: the partner
The partner is preparing the client presentation. Retained firms are explicit that resumes are insufficient on their own, that they weight behavioural evidence over pedigree, and that they warn clients against pattern-matching on titles at competitors. In practice the partner is reading for four things, in this order.
- Scope. What you actually owned: revenue, P&L, headcount, geography, budget, balance sheet. Stated outright.
- Results with the lever. What moved, by how much, and what you did to move it. A result without a lever is a claim. A lever without a result is a job description.
- Company-type signal. Public, private, sponsor-owned, founder-led. Each reads differently, and a partner who cannot tell which you have done cannot place you.
- Trajectory. Whether the career makes the next role the obvious step. This is the summary's job and it is the hardest part of the document to write.
The career highlights are the lines a partner lifts straight into the client presentation. Four or five, each fully bold, each with a number, each a result and the lever that produced it. If those lines are not on page one, the partner writes their own version of you, and it will be less precise than yours.
Reader three: the client
By the time the client sees your resume it sits inside a dossier with the partner's assessment, reference notes, and a comparison against three or four other candidates. The client is reading for fit against a spec and for the story: why this career leads to this seat. The mandate paragraph under each role does that work. Three or four sentences: why you were brought in, what you found, what you decided, how it unfolded. A list of bullets with no story reads as a job description, and the client has read four of those already that morning.
This is also where the company-type signal pays off. Boards favour the proven CEO when the situation is a reset: 41% of S&P 500 CEO appointments in the first quarter of 2026 had done the job before. Sixty-five percent of sponsors report replacing the CEO during a hold. A client in either situation is reading your resume for evidence you have operated in theirs.
AI in the loop
Two-thirds of search firms now use AI tools in their workflow, and 58% of US hiring managers use AI to screen resumes. Search-firm screening tools use language models that read for context. In the same surveys, 88% of hiring managers say they can tell when application materials were written by AI, and of those who care, a quarter read it as lack of effort.
Generic copy now fails both readers for the same reason: it has no specific content. "Results-driven leader with a proven track record" means nothing to a language model and nothing to a partner. A specific result with a named lever means the same thing to both. The way to pass AI screening and human screening is the same: write things that are true, specific, and only yours.
What to do with this
Build page one as a single argument a researcher, a partner, and a client can each read alone. Put scope where the researcher looks for it. Put the proof where the partner lifts it. Put the story where the client reads it. Then make LinkedIn say exactly the same thing, because that is where the search found you in the first place, and the partner will cross-reference both before your name goes into the deck.
And start outreach to search partners in week one of a search, before the document feels perfect. A retained firm builds its long list by mapping the market, so a partner may know your name before you know the role exists. The resume's job is to be ready when the partner asks for it.
Your resume is read after you are on the list.
By someone building a case for you. Give them the evidence in the order they need it.
