Real Data

What We Found Scoring Hundreds of Real CVs

We ran our CV scoring model against every real CV in our system, not some hand-picked set, and this is what came out of it: the gaps that show up most often, and how scores are spread out across real users.

How This Was Measured

442 of the 700 people who've signed up (63%) have a CV stored from onboarding, either uploaded directly or imported from LinkedIn, and we scored every one of them with the same model that runs our free CV Score tool: eight checks covering structure, contact details, quantified achievements, skills, phrasing, and summary quality.

This isn't a sample of random visitors, it's people who actually created an account, so read these numbers as what real users' starting CVs look like before they touch any of our tools, not as a survey of the general job market. One check, format and layout detection (Canva files, multi-column layouts), needs a signal we only capture in real time, and we've only got it stored for 7 of the 442 CVs, so we left it out of the numbers below rather than publish something built on 7 data points.

Score Distribution

Under 50, needs real work3.8%
50 to 79, workable with fixable gaps69.5%
80 and up, solid26.7%

Figures as of 16 August 2026, based on 442 CVs.

By CV Language

Scores vary by CV language, with a median of 79.5 for English-language CVs, 75 for Spanish, and 73 for French, based on the checks we can measure consistently across languages.

The Gaps That Show Up Most

Ranked by how often each check failed, across all 442 CVs:

Quantified results85.7%

Most CVs don't back up at least two of their three most recent roles with a number, a percentage, a headcount, a budget, anything measurable.

Contact completeness69.0%

Most are missing a LinkedIn URL or another basic contact field.

Summary quality50.2%

Half have no summary at all, or one that's too short to say anything.

Skills quality39.4%

Two in five list too few skills for the section to read as complete, or so many it stops looking curated.

Banned phrases22.4%

About one in five use at least one phrase ATS systems and recruiters have learned to filter out, things like results-driven or team player.

Section structure19.7%

About one in five are missing a full section: contact, summary, experience, education, or skills.

Experience descriptions9.7%

About one in ten have at least one role with no real description under it.

These numbers cover users who signed up and stored a CV with us, not the general public, and they'll shift as more people join. Last updated 16 August 2026.

See Where Your Own CV Lands

Run your own CV through the same model and see the full 8 point breakdown.

Score My CV