How scores are calculated
Every number, and where it comes from.
Scoring is deterministic: the same CV, job and profile always give the same result. That is what makes re-ranking free and every score explainable.
1. Criterion scores
Each criterion gets a score from 0 to 100 with a note — for example Relevant experience 82: 4.5 years using Laravel and PHP against a minimum of 3.
2. Job match
The criterion scores are combined using the job's weights, after any weight moves requested by your hiring profile.
3. Preference fit
Each active rule in your hiring profile scores the candidate from 0 to 100. High-strength rules count three times as much as low-strength ones.
4. Overall score
overall = (100% − influence) × job match
+ influence × preference fit
+ bonuses (max 10) − penalties (max 15)
Influence is 25% unless you change it.
5. Missing must-haves
If any must-have is missing, the recommendation becomes Missing mandatory requirements and the score is capped (59 by default), so a candidate without an essential requirement cannot outrank those who have it. You can change or remove the cap in the hiring profile.
6. Recommendation
| Label | When |
|---|---|
| Strong match | At or above the shortlist threshold (80) |
| Good match | A little below it |
| Needs review | At or above the review threshold (55) |
| Weak match | Below the review threshold |
| Missing mandatory requirements | Any must-have missing |
7. Rank
Candidates are ranked by overall score, then by how many must-haves they meet, then by job match.
Deep analysis
A deep analysis lets an AI model re-read the CV and adjust criterion scores and the must-have table. The overall score is still calculated by the same formula, so it stays comparable with everyone else's.