MERITUM

Transparency

How the Meritum Score works

A score nobody can inspect is a score nobody should trust. Every weight in the model is published here.

Model MERITUM-SCORE-V1.0

Content Quality

20%

Depth, originality and durability of published work across platforms.

Engagement Quality

20%

Meaningful responses and discussion, not raw impression counts.

Consistency

15%

Sustained contribution rhythm over time rather than isolated spikes.

Audience Growth

10%

Organic audience development. Deliberately a minor factor.

Community Contribution

15%

Replies, support and participation in other creators' work.

Cross-Platform Impact

10%

Contribution recognized across several connected networks.

Authenticity

10%

Signal integrity as assessed by the quality engine.

Design principles

  • Contribution over popularity. Follower count is deliberately a minor input; audience growth is capped at 10% of the model.
  • Determinism. The same inputs always produce the same score. No hidden randomness, no model drift between page loads.
  • Honest gaps. Platforms expose different data. Where a metric is unavailable, Meritum labels it as unavailable rather than estimating it.
  • Attributability. Every point traces back to a dimension, a weight and an underlying set of contribution events.

Known limitations of this proof of concept

  • Demo mode generates stable synthetic data when platform credentials are absent.
  • The quality engine is heuristic and is not production-grade fraud detection.
  • No MTI is minted, transferred or claimable; reward figures are illustrative.
  • Scores are not comparable to any live Meritum deployment.