---
title: "The five components and their weights"
description: "A TrackForge rating has five components: registration, collection, title, identity and risk. Each is scored per work from evidence, and the composite is a weighted mean, registration weighing most and risk least."
image: "https://trackforge.studio/docs/img/trackforge-social-card.png"
---

> Documentation Index
> Fetch the complete documentation index at: https://trackforge.studio/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# The five components and their weights

:::caution[Draft rule book, not yet calibrated]
The weights below have not changed since the method's first version. The coverage minimums, defect factors and floors are starting values in the draft rule book v2.4; the [method statement](/docs/certification/rating/method-statement) gives them.
:::

Each work is surveyed on five components. Each component asks one question of the work, and the questions are the five elements of a [property survey](/docs/certification/rating/what-a-rating-is).

| Component | Survey element | What is checked |
|---|---|---|
| Registration | Services connected | Registered at every in-scope collecting route where the work is used, with the right shares and parties, judged per route |
| Collection | Rent roll | Income consistent with usage; unmatched usage and unclaimed pools; income reaching the wrong payee. Needs statements |
| Title | Land registry entry | Ownership documented (agreements; US Copyright Office registration for US works) and consistent with what societies hold in the owner's name |
| Identity | Deeds and title plan | One ISRC per recording, one ISWC per work, writers with IPI numbers, recordings linked to the right work, no duplicates or collisions |
| Risk | Structural and legal defects | Competing claims and disputes, duplicate shells at a society, sources that disagree. Read only at PRS, MCPS and The MLC |

Registration carries the most weight, then collection, title and identity; risk carries the least. The composite is the weighted mean of the component scores over the components that were assessed.

## Risk is sound only where a risk check was read

PRS, MCPS and The MLC can show a dispute or competing claim; PPL and SoundExchange cannot. A work is sound for risk when at least one of PRS, MCPS or The MLC read it and raised no risk; a work none of them read is not checked, reason not supplied. Sound on risk means no known conflict was found in what was read, not that no risk exists.

## How a component is scored

For each component a work is in one of four states:

- **Sound**: the evidence shows the component in order.
- **Defective**: the evidence shows a named defect. The work's weight is deducted from the component in proportion to the defect's class: some classes count in full, some in part, and a share shortfall counts by the size of the gap.
- **Not checked** (the rule book's `unverified`): no source could answer, for one of six recorded reasons, each sitting with the client, the society or TrackForge: not supplied; too recent to appear; waiting for the society to publish it; not reached in this run; source could not be read; reading past its age limit. The [method statement](/docs/certification/rating/method-statement#why-a-work-is-not-checked) says who each sits with. A not-checked work is left out of both the numerator and the denominator and is reported as a coverage shortfall instead.
- **Not relevant** (`not_applicable`): the question does not arise for this work, for example a rights position the owner does not hold. This is not a defect.

A component's score is the sound weight divided by the assessed weight, so unverified works never flatter or punish the score; they show up as [coverage](/docs/certification/rating/the-tuple#coverage) instead. Unverified is only ever a recorded source limit, never a choice.

A work's overall condition follows its components: any defective component makes the work defective; otherwise any unverified component makes it unverified; otherwise it is sound.

## Weights across the catalogue

Each work carries a weight in the portfolio: mostly from how much the work is used in each channel (digital mechanical, physical mechanical, performance, neighbouring rights, sync), measured by statement lines, then Spotify lifetime plays, then YouTube views (converted into Spotify-equivalent plays where a channel mixes the two), and partly shared equally by every work, all scaled by the owner's attributable share of the work, so a small stake in a hit counts as a small stake. Observed income never sets a weight on its own: usage is the floor, and income can only raise a weight. A work with no usage evidence in any channel carries only its equal share and is reported in the unweighted tail. The [method statement](/docs/certification/rating/method-statement#how-works-are-weighted) has the detail.

## Collection is optional, and says so

Collection needs statements. Until the owner supplies them, the component is reported as "collection not assessed" and the composite is taken over the other four. A cap on the letter of a catalogue whose income has not been assessed is planned and is not yet in the rule book.

## Floors

Some defects are too serious for a weighted mean to express. The rule book carries structural floors: when a named share of the catalogue, by count or by income weight, carries a given class of defect, the letter is capped whatever the composite says. A floor that reads a route's income weight is evaluated once per route in scope and never applied at a route that is immaterial to the catalogue's income. Every floor applied is disclosed in the [tuple](/docs/certification/rating/the-tuple). The floors and their caps are listed in the [method statement](/docs/certification/rating/method-statement#from-scores-to-a-letter).

Source: https://trackforge.studio/docs/certification/rating/components/index.mdx