What recruiters actually do with your CV (an insider's view)

You hit "Apply". The CV goes into some void. 3 days later you get a rejection email — or more often, silence.

What actually happened in those 3 days? Based on conversations with 30+ recruiters across tech, finance, and design — here's the real workflow.

The first 10 seconds: the skim

Your CV hits the recruiter's inbox (or the ATS queue). For the first pass, the recruiter has maybe 20-40 CVs to triage per role, per day. They're not reading. They're scanning.

What they scan for, in order:

1. Company pattern — does the candidate come from similar-stage / similar-space companies?
2. Role fit — does the title match the JD's level (Senior / Staff / Lead)?
3. Top 2-3 bullets of the most recent job — is there evidence or not?
4. Tech stack / tool match — do the keywords from the JD appear on the CV at all?

This is a 10-15 second read. If none of these pattern-match, the CV goes into "no" pile. If at least one signal is strong, it goes into "maybe".

What kills you in the first 10 seconds

• Summary that says nothing. "Results-driven professional with strong communication skills" — the recruiter's eye slides right past this. Useless.
• Top 2 bullets of the most recent job describe responsibilities, not outcomes. "Worked on backend systems" vs. "Reduced p99 latency by 40% by…" — these read completely differently.
• No visible keyword match. If the JD wants "Python" and your CV says "backend development" without ever saying "Python", you look like a no.
• Titles don't match the JD's level. You're applying to a Staff role but your most recent title is "Software Engineer II" — unless the bullets clearly show Staff-level scope, you get filed as Senior.

The 60-second read (if you survived the skim)

If you're in the "maybe" pile, the recruiter will come back and actually read. This is where the remaining 80% of your CV matters:

• Bullets 3-5 of recent roles (do they back up bullets 1-2?)
• Projects section (especially for engineering / design / data roles)
• Education (only if it's a hard filter — Ivy/IIT/FAANG)
• Skills section (for keyword match, not quality signal)

At this stage the recruiter is forming the "is this a strong candidate" judgement. If the bullets all hold up under the "so what?" test, you move to the next stage.

What goes into the ATS note

Most ATS platforms let recruiters add a short note next to each candidate. These notes are what the hiring manager sees in the next round. Typical notes:

• ✅ "Strong backend, shipped at scale, has cover-letter stakeholder alignment story."
• ⚠️ "Experience fits but resume is weak — check if it's a writing problem or a real gap."
• ❌ "No Postgres experience. Passing."

Your CV's real job is to earn that ✅ — which means giving the recruiter specific things to write. Vague bullets don't generate specific notes.

What happens to the "no" pile

Depends on the company. At a serious one:

• Rejected within 1-3 days with an automated email
• CV stays searchable in the ATS for future roles ("silver medalist" searches)
• Never actually destroyed

At a lazy one:

• Sits in the queue indefinitely
• You get no response ever
• Recruiter closes the req when the role is filled

The hiring manager review (if recruiter says yes)

The recruiter forwards ~5-10 candidates to the hiring manager. The HM spends about 3-5 minutes per CV on this pass — they're forming the interview strategy.

At this stage, the HM is specifically looking for:

• Technical depth — do the bullets suggest the candidate can do the job, or just talk about it?
• Team fit signals — team size, seniority, scope of ownership
• Red flags — one-year stints, unexplained gaps, title inflation

What you can't do at this stage: change anything about your CV. What you can do: prepare answers for the specific things on your CV that might raise questions.

Why does the process seem so random?

Because it is variable:

• Recruiters vary wildly in how carefully they skim. Some are rigorous; some are exhausted.
• Time-of-day matters. First CV of the day gets more attention than #40.
• Ordering matters. Your CV relative to the other CVs in the batch affects outcomes.
• A personal referral at the company bumps you past most of the above.

The best way to insulate yourself from randomness is to make every CV strong enough that even a tired recruiter at 5pm gets the right signal in the first 10 seconds.

What this means for how you write your CV

1. Front-load the signal. Assume the recruiter reads 2 bullets per role and stops. Put your strongest evidence first in each role's bullet list.
2. Make keyword match obvious. Not a comma-separated wall — inline, in the bullets. "Designed a FastAPI service in Python on AWS ECS" > "Skills: Python, FastAPI, AWS."
3. Write for both passes. Strong top-of-page for the skim; dense-but-readable body for the 60-second read.
4. Tailor per role. The skim is primarily about pattern-match with the JD. A generic CV fails the skim.
5. Follow up. If you haven't heard back in 7 days, a polite LinkedIn note to the recruiter moves you from "no" to "maybe" surprisingly often.

How CVCL thinks about this

CVCL's output is deliberately designed for the skim. The top of the CV (summary + first role) is where we're most aggressive about keyword insertion and quantified outcomes. The bullets get tighter and more detailed as you read further down — matching the attention curve of the actual reader.

This is why a CVCL-generated CV looks different from what you'd produce by hand in a template: we're optimising for a specific reading pattern, not for visual prettiness.

3 free generations on signup at cvcl.online.

TL;DR

• Your CV gets a 10-second skim, first
• If it passes, it gets a 60-second read
• The recruiter writes a short ATS note. Your CV's real job is to generate a specific, positive note.
• Top-of-page + front-loaded bullets matter more than total content volume
• Randomness is real. Follow up after 7 days.