Scoring explanation and published rating outputs

Good On You’s methodology point-scoring system is translated into an easily comprehensible five-point scale with labels ranging from “We Avoid” to “Great”. This article explains what’s included in published ratings, and how the scoring system works: its weighting, conditionality, application to the rating labels, and more.

Rating outputs

The key pieces of information that form a published rating are detailed below.

Pillar and overall ratings

The overall and pillar ratings simplify Good On You's comprehensive assessments into a five-point scale and associated rating label.

Good On You’s published ratings feature the rating label for each pillar (“planet”, “people”, and “animals”), and an overall rating. These both use the same five-point scale and labelling.

• “Great” (5/5): Brands rated “Great” demonstrate leadership in all three pillars. They are typically very transparent and have both strong policies and strong assurance (for example, from relevant certifications or standards systems) to address the most significant issues across their supply chain.

• “Good” (4/5): These brands adopt policies and practices to manage multiple issues across their supply chain and often demonstrate leadership in one or more areas.

• “It's a Start” (3/5): These brands are transparent about their policies and practices to manage some issues and are making good progress on one or more of them.

• “Not Good Enough” (2/5): These brands disclose some information in one or more areas and consider some issues but do not yet sufficiently address the impacts across their supply chain.

• “We Avoid” (1/5): These brands disclose little to no relevant or concrete information about their sustainability practices. In some cases, the brand may make ambiguous claims that are unlikely to have any positive real-world impact.

See the score to label translation below

The methodology and its scoring are constructed such that “Good” and “Great”-rated brands are considered the leaders in sustainability action in a particular industry vertical, worthy of promotion as more sustainable options for consumers and business promotion.

Rating summary

The text-based summary published with a rating collates the most important points that explain the brand's rating in each pillar. It is a constrained summary of the practices most indicative of the brand's activity in that pillar, and thus is a very small subset of the total activities assessed in the rating. 

When the rating data assessment is completed, the analyst signing off on the rating will generate an “auto-summary” using a logic-based algorithm that selects the most important brand practices to explain the rating across each pillar and transforms it into a text-based summary. The analyst refines the summary ready for publication.

Values and certifications

Ratings that are 3 out of 5 or higher are published with some select badges for values that represent the brand's sustainability priorities and certifications it prioritises, for example valuing organic materials or Fair Wear Foundation membership.

See the values and certifications definitions here.

Last rated date

The “last rated” date published with the rating means the date the rating was finalised and the disclosures were verified as valid support for the rating data.

Scores

The normalised overall and theme-level scores out of 100 are generated based on the scoring logic processing the full rating data. Those scores are published in Good On You enterprise data products and in Good Measures. Item-level scores are published in Good Measures and will soon be available in enterprise data products.

Full rating data

The underlying question-and-answer data, including evidencing URLs, is available to brands in Good Measures and data services customers via bespoke data requests. This data will soon be available in enterprise data products.

Scoring explanation

The basic scoring logic is as follows:

• The methodology is a series of questions with fixed question options and/or percentage-proportion answers

• The methodology hierarchy is: pillars > areas > issues > questions > question options (QO)

• Each question option (QO) has a score ranging from 0-100

• Each question has a point weighting or "achievable points" (ie 5, 10, 15)

• The score on an issue, area, and pillar is calculated by adding the points scored on each question within that section (score numerator), divided by the total achievable points for those questions (score denominator).

• The score is normalised as a percentage. A brand can earn between 0-100% of the achievable points on each issue, area and pillar relevant to their rating.

• The brand’s overall score is a simple average of its scores on each pillar.

• The overall and pillar rating out of 5 is a mapping of score bands to each of the five rating labels (e.g. 90-100% = Great)

Question scoring

Each question and available answer has an assigned number of points that can be achieved. This is based on the weighting determined through our research and consultation processes. 

Question formats are: boolean (yes/no), single choice, multiple choice, or percentage or distribution (ie multiple-choice question options with an associated percentage input, for example: organic cotton, 44%). 

For multiple-choice and distribution questions, points are awarded cumulatively, per each data point provided, up to the maximum points available for that question. Other questions allow only one selection. There are no open-ended questions with associated scores in any methodology.

An answer adds to the score of the issue, area, and pillar it is contained in by multiplying the question option score by the question's achievable points. For example, if a question is worth 2 points for the Planet, and the selected answer is worth 50% for that question, then the brand will score 1 point for Planet.

The answer to a question may impact the score of another

An additional method of scoring is when the answer to one question is relevant to another. Selecting an answer may trigger the scoring system to score a defined number of points in another section of the methodology. Answers to this second section will still be recorded and rewarded in addition to the scoring applied through this factor. 

For example, a brand’s use of organic materials will inherently reduce its chemical use. Selecting an answer about the use of lower-impact materials may trigger the scoring system to apply a defined number of points for the chemicals score. Brands’ chemical reduction practices additional to the use of lower-impact materials will still be recorded and rewarded under the relevant chemicals section.

Materiality: the weight of an issue may differ for certain brands 

The scoring applied by the Good On You methodology seeks to reward brands for good sustainability practice, taking into account the materiality of the sustainability issue being addressed and the relative impact of the particular practice they have adopted. This means:

1. Different sustainability issues have different score weights (reducing greenhouse gas emissions is weighted much more highly than reducing packaging impacts, for example).

2. Issue weightings vary according to how material they are in a specific industry (packaging in the beauty industry is relatively more impactful than packaging in the fashion industry).

3. Different practices deserve greater or lesser reward (eliminating packaging altogether v using recycled materials in packaging)

Score weighting for the same question can differ depending on a brand’s size and its product range. Here are some illustrations of how:

1. Brand size: For example, both small and large brands are assessed on the “water” area, but the relative weight is higher for large brands as they have more resources and influence over their supply chains.

2. Product range: For example, scoring weight may differ according to the proportion of particular materials (fashion vertical) or commodities (beauty vertical) that a brand uses. Such as:

   1. Microplastics: The weighting of this issue is commensurate with the proportion of synthetic fibres used in the product range.

   2. Deforestation: The weighting of this issue is commensurate with the proportion of forest-derived materials (leather, rubber, metal, precious stones, and cellulosic fibres) used in the product range.

   3. Leather tanning: Weighting can vary depending on whether leather is a dominant or non-dominant material in a brand’s product range. If a brand does not sell leather products the weighting is zero. The issue and its questions are subsequently disabled for the rating. 

   4. Solvent use: Do you sell shoes? The weighting of the issue is zero if the brand does not sell shoes (the issue is disabled for the rating and questions are not answered), and can vary depending on whether shoes are a dominant or non-dominant category.

Overall score and label calculation

Brands are allocated a score out of 100 for each rating pillar: planet, people, and animals. 

The overall score out of 100 is calculated as the average of those pillar scores. 

Summary labels (“We Avoid”, “Not Good Enough”, “It's a Start”, “Good”, or “Great”) are banded from the overall score ranges detailed in the table below.

Brands in categories that do not commonly use animal-derived materials (eg those that only offer swimwear) are not scored on the animals pillar, and instead their score is an average of the people and planet scores. However, in the beauty methodology, all brands are scored for animals. This is because animal welfare concerns are more wide-ranging for beauty products: animal-derived ingredients and animal testing are issues across all product categories.

Scoring and normalisation of conditionality

Good On You’s methodology questionnaire contains multiple routes depending on features specific to a brand, so not all brands will have the same questions in their assessment. 

For example, only a brand that uses leather will be assessed on leather tanning questions in the chemicals section. The selection of question routes impacts the number of questions presented for the brand, and therefore the denominator of their score will change. Brands that have engaged in more sustainability initiatives will generally have a higher denominator compared to those that don’t, as question answers are often “no”. 

Scores are calculated as a percentage to normalise the effect of different conditional question routes. The final score is the number of points awarded divided by the total number of points available to the brand for the question routes selected. 

Variance in score numerators and denominators is normalised out of 100 to provide a comparable score metric across all ratings. If a brand has fewer answered questions because some questions are not relevant to that company, it is not penalised or placed at a disadvantage.

Substitutions: 3rd party standards (3PS) and business practices inform scores in some issues

In several questions throughout the methodology, brands can receive scores for certain issues that are already covered through the following mechanisms:

1. 3rd party standards (3PS): this means certifications, standards, ratings from other organisations, and multi-stakeholder initiatives (MSI). For example: a brand with the Carbon Trust Standard (Waste) certification substitutes points in the non-textile waste section.

2. Business practices: this refers to ways that a brand operates which directly mitigate issues within the methodology. For example: operating in lower-medium risk (for labour abuse) countries means brands are substituted some points in particular labour issues, such as modern slavery and Code of Conduct.

Note that these are not bonus points. A brand can score a maximum 100% of the achievable points on each issue through a combination of selecting answers within the item and through substitutions stemming from answers in other issues. 

Question clusters

There may be some questions that essentially share the same core issue, and therefore have the same achievable total score.

For example, we ask for the percentage of recycled packaging and other packaging initiatives in separate questions for easier comprehension, but they ultimately count towards the same total achievable points.

Pillar caps, floors, and ad-hoc rules

Pillar scores can be capped at a certain level if particular practices disqualify the brand for a higher rating. 

In some cases where brands have a lower impact than other brands by their nature, a score floor ensures they do not score below a certain level on a pillar. 

There are other scoring rules to account for special cases. For example:

1. Major incidents cap: Brands that have had major incidents related to people, planet, and animals in the three years before rating, and took little action to remediate, have their score capped at 64 ("It's a Start" rating) for that area or 44 ("Not Good Enough" rating) in the case of multiple instances. This cap applies only to large brands. It doesn't apply any limit to the overall score and rating. The cap applies across fashion and beauty verticals.

2. Animal products cap: If a brand uses animal products, the animal pillar rating is capped at 84 ("Good" rating). While the methodology recognises and scores positive practices related to the use of animal materials, we also acknowledge that alternative non-animal-derived materials exist that are preferred for minimising risks of harm to animals. Rating "Great" in animals means brands are seeking out alternatives to eliminate animal product usage. This cap doesn't limit the overall score and rating: a brand can still use animal products and score "Great" overall. This applies to the fashion vertical.

3. Small brands planet floor: Small brands have a floor of 15/100 on the planet pillar score. This makes large and small brand methodologies more directly comparable since small brands inherently have limited impact across all environmental issues by virtue of their smaller scale operations.