TrueCalling glossary

What is TrueFit 360? Candidate fit scoring, explained

TrueFit 360 is the proprietary candidate-job fit score used by TrueCalling that combines four explainable axes (technical skills, experience context, soft signals, and intent level) into a single auditable rating out of 100, with every contribution traceable to a public source signal. It is the candidate fit scoring method behind TrueCalling shortlists.

Definition reviewed on

How does TrueFit 360 score a candidate?

  1. What does the fit score measure?

    TrueFit 360 compares a candidate's CV or profile with the job description and returns a compatibility score from 0 to 100. The score arrives with the skills that match, the skills that are missing, and the gaps, split into major gaps (required skills that are absent or held for under two years) and minor gaps (nice-to-have skills). One rule keeps the number honest: a candidate with three or more major gaps cannot score above 50 (TrueCalling internal data, 2026). Each score also carries a recommendation (strong yes, yes, maybe or no) and a two-to-three-sentence summary in the recruiter's language. The limit is the evidence: the score reads only what the CV or profile says, so a skill the candidate never wrote down stays invisible, and a low score on a thin profile means "not shown" rather than "not able".

  2. Which secondary signals sit next to the score?

    TrueFit 360 adds two secondary scores from 0 to 100, computed by fixed rules rather than by the language model, so the same CV always gets the same result. Professional resilience reads the career: the number of employers, stints shorter than 12 months and progress towards senior titles. Digital reputation looks for public professional evidence in the CV, such as a LinkedIn or GitHub link, certifications, publications or conference talks, and starts at 40 when there is none (TrueCalling internal data, 2026). Intent is read later, from the conversation: EMILY™ by TrueCalling classifies each WhatsApp reply as interested, not interested, unclear or unsubscribe. The limit: both rule-based scores reward visible signals, so short contracts and quiet profiles score lower. Read them as context, never as a filter; our guide to AI bias in recruiting shows how to test for that effect.

  3. How can a recruiter check or challenge a score?

    Every TrueFit 360 score can be opened and questioned by the recruiter or the hiring manager. A "Why this score?" action writes an explanation of four to six sentences that ties the number to specific experience and skills in the CV. Behind the screen, each AI scoring decision is logged with the exact model version, a prompt identifier, the input and the output, and kept for 6 months by default (TrueCalling internal data, 2026). The company also chooses which AI provider runs the scoring. That log supports the record-keeping duty in Article 12 of the EU AI Act, covered in our AI Act recruiting compliance guide. The limit: an explanation shows how the model reached its number, not that the number is right, so recruiters should check it against the CV before passing it on. For how scores work across tools, read the candidate matching score guide.

How does TrueFit 360 differ from a single match percentage?

Legacy approachTrueFit 360 (TrueCalling)
One match percentage with no breakdown.Score 0–100 with matched skills, missing skills and ranked gaps.
Missing required skills averaged away.Three or more major gaps cap the score at 50.
No reason given for the number.Written summary per score; a longer explanation on demand.
No record of the model behind a score.Model version and prompt logged per score, kept 6 months.

FAQ: TrueFit 360

What is a good TrueFit 360 score?

There is no universal pass mark. A TrueFit 360 score is relative to one job description, so compare candidates for the same role, not across roles. Read the gaps with the number: two candidates on 75 can differ, one missing a required skill and the other only a nice-to-have.

Which data does TrueFit 360 use to score a candidate?

The candidate's CV or profile and the job description. The compatibility score compares the two, the resilience score reads the career history, and the digital reputation score looks for public professional links, certifications, publications and talks mentioned in the CV. WhatsApp replies are classified separately, for intent.

Can a company choose the AI model behind the score?

Yes. The AI provider is a company-level setting in TrueCalling, and the model version used for each score is written to the audit log. Any score can therefore be traced to the model that produced it, even after the company changes provider.

How long does TrueCalling keep the scoring log?

Six months by default. Each AI scoring decision is stored with its model version, prompt identifier, input and output, then deleted by a daily job once its retention date has passed. The retention date is stored per record, so it can be extended when an audit requires it.

See TrueFit 360 in action

30-minute demo of TrueCalling against one of your open roles: sourcing, scoring, and outreach end-to-end.