Scoring Methodology

This page explains how every HIP Score is produced. Transparency about methodology is a core part of what makes a rating credible — and it is our strongest defence if a score is ever questioned.

HIP Ratings are commissioned and paid for by the manufacturer of the product assessed. Scores are produced solely by applying our published methodology, with human review, and the fee does not influence the result.

Current version: 1.5 · Effective July 2026

Download methodology.yaml (v1.5)
Download full methodology (PDF, v1.5)

The full methodology document sets out each scoring dimension, the Regenerative Index tiers, category ceilings, and — under standards alignment — the ISO clause-mapping audit trail that underpins HIP Score version control.

What the HIP Score measures

The HIP Score (Human Impact Profile Score) is a material resilience and sustainability rating, not a carbon footprint or general environmental endorsement. It assesses seven dimensions related to materials, supply chain, and end-of-life characteristics. See the About page for a full summary.

How the score is calculated

Step 1 — Score each dimension independently

Each of the seven dimensions is assessed separately using a structured scoring rubric — a set of written criteria for each score band (0–2, 3–4, 5–6, 7–8, 9–10). The rubric for each dimension is defined in methodology.yaml (link below).

Step 2 — Normalise the Regenerative Index

The RI uses a scale of −10 to +10. Before weighting, it is normalised to 0–10 using:

RI Normalised = (RI Raw + 10) ÷ 2

This maps −10 → 0, 0 → 5, +10 → 10.

Step 3 — Calculate the weighted HIP Score

HIP Score = (MSI × 0.20) + (SCR × 0.18) + (RC × 0.18) + (R × 0.13) + (SEI × 0.08) + (PL × 0.08) + (RI_norm × 0.15)

The result is rounded to one decimal place.

Dimension weights

Category-specific weight profiles — As of v1.5 (July 2026), dimension weights vary by product type. The base weights below apply when no category-specific profile is matched. Seven product-type profiles are defined: buy-it-for-life durables, consumable devices, consumable units, safety-sealed products, software-dependent electronics, battery-bearing electronics, and single-use packaging. See the methodology source file for the full weight table.

* Normalised to 0–10 before weighting using the formula above.

HIP Mark tiers

The HIP Mark is awarded when the HIP Score meets a minimum threshold:

Products scoring below 6.0 receive a HIP Score on the rating page but do not carry the HIP Mark.

Manufacturers holding a current verified rating and signed licence may display the HIP Mark. See the Brand Guidelines for visual standards, permitted usage, and licensing requirements.

Rating types: Baseline, Screened, and Verified

Category Baseline ratings are produced using publicly available data only. They rate a product category — not a specific named product — and apply conservative assumptions wherever data is missing. The result is an honest baseline for the category.

Screened Ratings are product-specific assessments using publicly available data plus a structured manufacturer submission. They do not require a full evidence dossier. A Screened Rating replaces the conservative defaults of the Category Baseline with information about the specific product, so the resulting score reflects the real product — which may be higher or lower than the category default. Commissioning a rating buys an accurate assessment, not a higher score.

Verified ratings are produced using data submitted by the manufacturer through the Resourcehip submission form. Submitted data replaces the conservative category defaults. Scores above 5 on the Social & Environmental Impact dimension require third-party audit evidence — self-reported claims alone cannot exceed 5.

Where a verified rating uses a category default for a dimension (because no specific data was submitted for that dimension), this is clearly noted as an improvement opportunity on the rating page.

The AI scoring pipeline

Ratings are produced using a local AI pipeline running on Resourcehip's own hardware using Ollama. No product data is sent to external cloud AI services.

No rating is ever published automatically without human sign-off.

Data sources

The following public data sources are used across ratings. Each rating page lists the specific sources used for that assessment.

All data sources are public domain or published under open data licences. Attribution is shown on each rating page.

Conservative by design

Where data is missing, the pipeline uses the most conservative reasonable assumption — not the most optimistic. This protects the integrity of the rating and creates a genuine incentive for manufacturers to submit better data. A verified rating with real data replaces those conservative defaults, so the resulting score reflects the actual product rather than worst-case assumptions — it may be higher or lower than the category default.

Independence and conflict of interest

Resourcehip charges manufacturers a fee to produce verified ratings. This creates an obvious question: does paying for an assessment buy a better score?

The answer is structural, not aspirational.

The fee buys an assessment, not an outcome. The assessment fee covers the cost of reviewing a manufacturer's evidence dossier against the HIP methodology. It does not purchase a minimum score, HIP Mark eligibility, or favourable treatment. The scoring pipeline applies the same rubrics, weights, and conservative defaults regardless of whether a product was submitted by a manufacturer or rated as a Category Baseline from public data.

Sub-threshold results are published. If a verified assessment produces a HIP Score below 6.0 — the minimum for a HIP Mark — or below the category baseline, that result is published on resourcehip.com. Manufacturers cannot withdraw from publication after submitting data for assessment. Suppressing unfavourable results would destroy the mark's value for every manufacturer who earns it honestly.

Structural scoring limits apply regardless of payment. The methodology enforces caps that no commercial relationship can override:

• Social and Environmental Impact (SEI): Self-reported claims cannot score above 5 out of 10. Scores above 5 require submitted evidence of third-party audit or certification — ISO 14001, SA8000, B Corp, or equivalent.
• Regenerative Index (RI): Self-reported claims cap at +4. Scores above +5 require third-party verified evidence — ROC, FSC, PEFC, or equivalent.

These limits are enforced in the scoring pipeline. An assessor cannot override them.

Public reporting commitment. Resourcehip will publish, annually, the percentage of verified assessments that do not qualify for a HIP Mark. This figure is a basic test of methodology credibility: if every verified product qualifies, the bar is too low. We expect a meaningful proportion of products to fall below 6.0, and we will report that figure transparently.

How we review and update ratings

Ratings are reviewed by exception, not on a fixed calendar. Each week, the pipeline re-scores the full catalogue against the current methodology and public data sources. Only ratings where the score change is material — a shift of 0.5 HIP points or more, or a crossing of a HIP Mark tier boundary — enter the human review queue.

Ratings that do not change materially are automatically confirmed as current each quarter (the "last reviewed" date is refreshed). This means every published rating is either recently re-validated by a human reviewer or confirmed unchanged by the scoring pipeline within the last three months.

To guard against stale ratings that happen to remain numerically stable, Resourcehip also conducts periodic random-sample audits. Each quarter, approximately 5% of the catalogue — weighted toward commercially active categories (those with verified ratings or paid HIP Marks) — is pulled into the review queue regardless of whether the score changed. These audits are logged alongside standard reviews for full traceability.

A methodology version change (e.g., new dimension weights or rubric updates) triggers a full re-score and human review of every published rating, regardless of delta.

Methodology version history

In July 2026 we introduced category-specific weight profiles. Instead of applying the same dimension weights to every product, seven product-type profiles now tailor the weighting to what good performance actually looks like for that kind of product. A buy-it-for-life durable like cast iron cookware is weighted more heavily on repairability and longevity; a consumable-dependent device like a coffee pod machine is weighted more on resource circularity. Of our 73 published ratings, five scores moved (cast iron cookware +0.8, electric toothbrushes +0.4, robot vacuums +0.3, cordless drills +0.2, coffee pod machines −0.1) and none changed band. The full per-category deltas are published at V1.4 delta report.

Standards alignment

ISO 14024:2026 (Type I Ecolabels)

HIP is ISO 14024-informed, not ISO-certified. Resourcehip is building toward certification: Phase 2 (Q4 2026–2028) adds independent peer review, a Criteria Board, and third-party audit.

ISO 14024:2026 is the international standard for Type I environmental labels — independent, multi-criteria ecolabels based on lifecycle assessment. The HIP methodology was designed with these principles in mind.

What HIP aligns with:

What HIP does not yet claim:

The Regenerative Index (RI) is a proprietary Resourcehip extension — there is no direct equivalent in ISO 14024 — and reflects our view that material resilience must account for regenerative potential, not only harm reduction.

Phase 2 certification roadmap:

Licence and source

The HIP Score methodology is published under Creative Commons CC BY-NC 4.0 — free to share and adapt for non-commercial purposes with attribution to Resourcehip.

The authoritative source file is methodology.yaml (v1.5), maintained in the Resourcehip pipeline repository.

The HIP Mark logo and Resourcehip brand are proprietary and may not be used without a licence.

Dispute a rating

If you believe a rating contains an error, see the Dispute a Rating page for the review process.

Methodology v1.5 · Resourcehip Ltd · Terms · Privacy