Skip to content
PlasticFreeLab
12 in the lab47 queuedUpdated weekly

PFL Methodology v1.2

Specification · v1.2

Updated April 2026 · Previous version: v1.1 ↗ · All versions

This is the working specification for every product score published on PlasticFreeLab between April 2026 and the next versioned release. Five weighted dimensions, twenty-three sub-tests, four reviewer roles, and a dispute policy that doesn't move under pressure.

The five dimensions

Every score on this site is a weighted average of the five dimensions below. The weights reflect what we believe matters most for a non-toxic-living recommendation: material safety dominates because it's why our reader is here, and value sits last because price is recoverable while exposure is not.

  • Material safety (35%): What it's made of, what it leaches, what's certified.
  • Performance (20%): Does it actually do its job, well, under realistic load.
  • Durability (15%): How well it survives 24 months of normal use.
  • Use experience (15%): What it's like to live with for a month.
  • Value (15%): Cost per year of expected service life, relative to category.

Material safety · 35%

Material safety is what most readers come for and the dimension we weight highest. We score the disclosed bill of materials, third-party certifications (NSF, GOTS, OEKO-TEX, MADE SAFE, Greenguard), and any contaminant test data the manufacturer publishes. Where the manufacturer is silent, we order our own panels through accredited labs.

Sub-tests

  • PFAS panel (SGS Method 5021A): Quantitative analysis for 40 PFAS compounds at parts-per-billion sensitivity. New in v1.2.
  • Lead and cadmium screen: Surface-contact extraction per FDA Compliance Policy Guide 7117.06. Pass / numeric value reported.
  • Phthalate panel: GC-MS analysis for the eight phthalates restricted under CPSIA. Reported in micrograms per gram.
  • VOC chamber: 72-hour CDPH 01350 chamber test for total VOCs and the priority list of 35 individual compounds.
  • Supplier disclosure audit: Tier 1 manufacturer published, with country and certification of each material layer.

Performance · 20%

A non-toxic product that doesn't perform is not a recommendation. We measure category-appropriate function: heat distribution and retention for cookware, contaminant reduction and flow rate for filters, structural recovery and breathability for textiles. Performance scores are bench-measured, not vibe-checked.

Sub-tests

  • Infrared heat map: Surface temperature mapped at 30-second intervals across a 12-minute heat curve.
  • Deglaze + sear panel: Standardized protein sear plus acidic deglaze, scored on browning, release, and surface integrity.
  • Contaminant reduction: NSF-spec influent challenge water, effluent measured for the ten priority contaminants.
  • Load-cycle test: Full-capacity continuous use over the cartridge's stated lifetime, with end-of-life retest.

Durability · 15%

Coatings chip, elastic dies, gaskets shrink. Durability scoring captures the failure modes that turn a non-toxic product into a sub-toxic one twelve months in. Our panel runs each item through a 24-month accelerated wear cycle plus longitudinal field reports from real households.

Sub-tests

  • Accelerated coating wear: ASTM-style abrasion cycle equivalent to roughly 24 months of normal home use.
  • Warranty audit: Coverage, exclusions, and median claim resolution time pulled from public small-claims filings where available.
  • Longitudinal field reports: Six panel households report monthly for 12 months. Mid-test attrition is itself a signal.

Use experience · 15%

Weight, balance, cleanability, learning curve, sound, smell. The qualitative dimension, but scored against fixed rubrics so the result reproduces between reviewers. Three panel members each spend a calendar month with the product before scoring.

Sub-tests

  • Weight + ergonomics: Mass, balance, and reach for the relevant ergonomic dimensions. Scored against a category baseline.
  • Cleanability: Dishwasher and hand-wash cycles, scored on residue and finish degradation.
  • Panel qualitative scoring: Three reviewers, fixed rubric, blind to brand where the geometry allows it.

Value · 15%

We score value as expected cost per year of service over a 10-year horizon, normalized against the category median. A pricey item with a 30-year warranty often outscores a cheap item with a 2-year coating. Warranty terms, refurbishment availability, and resale market all feed in.

Sub-tests

  • Expected cost per year: Purchase price plus consumables, amortized over expected service life.
  • Warranty value adjustment: Multi-year warranties that pay out reduce the per-year cost; weak warranties don't.
  • Market comparable score: Indexed against the median in-category cost-per-year of all current shortlist entrants.

Lab partners

We use SGS, Eurofins, and Intertek for quantitative chemical panels. For PFAS specifically, SGS's Method 5021A is our default. Lab partners are listed as a transparency commitment; partner status doesn't affect scoring, and labs do not see brand identity until the report is filed. (Full partner list and reciprocity policy: to be wired alongside the launch of the v1.3 spec.)

Reviewer credentials

Four roles, each empowered to kill a sentence the team can't defend.

Research lead

Reads the primary literature. Cites the trial, not the press release. Disagreement with the consensus only shows up when the data earns it.

Testing lead

Cooks on every pan for a month. Weighs every bottle. Measures every claim. Tools and notebooks outnumber opinions.

Medical reviewer

Flags YMYL claims. Confirms no fearmongering made it past the edit. Quietly removes the scary sentence.

Sourcing lead

Reads spec sheets. Triangulates manufacturer claims with third-party test data. Names the brands we reject, and says why.

Dispute resolution

If a brand believes our scoring is wrong, they can submit a dispute with documentation. We re-test on our bench, not theirs. If their data contradicts ours, we publish both. We do not edit live scores under threat of legal action; we update them when better evidence arrives.

Change log

  • v1.2 (April 2026): Added PFAS panel via SGS Method 5021A. Expanded durability window from 12 to 24 months. Revised value formula to weight warranty terms.
  • v1.1 (January 2026): Initial public release. Five-dimension rubric, five reviewer roles, dispute policy.