Skip to content
← See all articles
Matching 7 min read

The matching score explained: how AI ranks your jobs

A “72%” with no explanation is useless. Here’s how to read a matching score, what it surfaces, and why it saves you hours.

MATCHINGThe matching scoreFrom noise to signal, explained

Job hunting isn’t about a lack of listings, it’s about having too many, and not knowing which ones deserve your time. A matching engine solves exactly that: it reads every job, compares it to your real profile, and gives it an explainable relevance score. Here’s how to read it, without giving away our recipe.

A score, but above all, an explanation

A number on its own is a black box: a “72%” with no justification helps no one. What matters is the why, what in your profile answers the job’s need, and the small gap worth knowing before you apply. So every score comes with its explanation, in plain language.

From noise to signal: what the ranking surfaces (orders of magnitude)Jobs found · 100In your field · 42Well ranked · 18Top matches · 6

Reading meaning, not keywords

Two jobs can describe the same role with completely different wording, a plain keyword comparison would miss it. The engine reads the meaning of both the job and your profile, across languages, to measure how close they really are. That’s what surfaces a good job even when it doesn’t use your exact words.

Your specialty comes first

The most telling signal: is the job actually aiming at your real profession? A nurse who also handles admin work shouldn’t see every secretarial role float to the top; a salesperson comfortable with a spreadsheet isn’t a financial analyst. The engine favors your specialty over a mere shared skill mentioned in the body of the ad, that’s what keeps “close but not quite” roles out of the top of the ranking, whatever your field.

Key takeaway : A good score isn’t the one that ticks the most boxes, it’s the one that reflects your real specialty. Exactly how we weigh each signal stays proprietary, but the result is always explained to you.

Several signals, one explained score

On top of meaning and specialty, other weighted signals come into play, how well your background fits the need, and practical constraints like location or contract type. They’re combined into a single score, then translated back into readable strengths and gaps. The exact recipe stays with us; what you see is a clear decision.

A transparent score is a decision tool, not a gimmick: it tells you where to put your energy.

What the score never does

It never rejects a job on location alone, and it never inflates your profile to improve a number. The ranking stays honest: it reflects your real background, not an idealized version.

Frequently asked questions

Can the score be wrong?
No score is perfect: it helps you prioritize, not decide for you. Every score is explained as strengths and gaps, you always have the final say.
Why does a job with my skills get an average score?
Often because your specialty isn’t in the title: the role shares tools with you but targets a different profession. The engine then ranks it as “close” rather than “on target”.
Does matching work outside of tech?
Yes. The engine is generic: healthcare, finance, consulting, engineering, marketing, HR, legal… it adapts to your field from your real CV.
Put this into practice

Generate a tailored CV and cover letter for your next job, grounded in your real experience.

Get started
Read next
AI
Job hunting with AI in 2026: the complete guide
CV
ATS-optimized CV: the complete guide to passing the filters (2026)
Cover letter
The cover letter that converts: a 5-part structure