How we test movie matches
A useful match preserves the part of the first film that mattered while clearly stating what changes. Genre overlap alone does not pass.
1. Build broad recall
Related-title data creates a candidate pool. Movie DNA compares atmosphere, emotion, pace, twists, humour, romance, scale and groundedness. Trust-adjusted quality is a supporting signal, not permission to recommend an irrelevant classic.
2. Human review the priority route
For every full Match Report, the editor checks one lead and three alternatives. Each needs a specific shared reason, a meaningful difference and a caution. Nonsensical explanations are rewritten; weak candidates are replaced.
3. Test different evenings
We add a couple option, family option where responsible, and a late-start option. A title can be a close artistic match and still be a poor situational choice.
4. Listen, then publish evidence carefully
Readers can mark “nailed it” or “not similar.” We use this to review the route. Public success percentages will appear only after enough independent responses exist; tiny samples are not presented as proof.
What we do not claim
Movie DNA is explainable judgement, not scientific truth. Streaming availability is a dated regional check. Scores, ads and affiliate relationships do not buy placement.