Methodology

How we score, and what we refuse to score

Version 1.0 is a desk comparison. Every figure comes from a page the manufacturer publishes, and the scoring is arithmetic you can repeat from the numbers printed on each page.

LowTox Review editorial team Published 14/08/2026 Method v1.0

What we score and why

A shopper standing in front of two boxes of dishwasher tablets cannot run a laboratory test, and neither can we yet. What that shopper can do is read what the brand publishes, and that turns out to separate the field more sharply than most people expect. Across the categories scored so far, eight of the seventeen products scored contain polyvinyl alcohol. Four of those say so on the product page and four do not.

So version 1 scores the paper trail: what the brand publishes, where it publishes it, whether the claims and the ingredient list agree, what a single use costs, what the packaging is made of, and which third-party marks the product actually carries. All five are checkable by anyone with a browser, which is the point. If you disagree with a score you can open the same page and argue with the band rather than with our taste.

The five criteria and their bands

The scoring rubric, version 1.0

Criterion	Weight	Bands
Ingredient transparency	25%	9 out of 10. Full ingredient list published on the product page.5 out of 10. Full list published by the brand, but not on the product page. A separate ingredients page or a help centre article.3 out of 10. Full list on the product page, contradicted by a claim on the same page.2 out of 10. No first-party list retrievable. The list comes from a third-party transcription.1 out of 10. No ingredient list published anywhere we could find.
PVA and microplastics	25%	10 out of 10. No polyvinyl alcohol and no microplastic-shedding format.5 out of 10. Contains polyvinyl alcohol, and the product page says so.2 out of 10. Contains polyvinyl alcohol, and the product page does not say so.
Cost per use	20%	9 out of 10. Within 10 per cent of the cheapest use in the category.7 out of 10. Up to 25 per cent dearer than the cheapest.5 out of 10. Up to 50 per cent dearer.3 out of 10. Up to twice the cheapest.2 out of 10. More than twice the cheapest.
Packaging	20%	10 out of 10. Plastic-free refill system: a durable container you keep, plus plastic-free refills.9 out of 10. Plastic-free pack: paper, card or metal, with no plastic component described.6 out of 10. Recyclable single-material pack that is either not plastic-free or not a refill format.3 out of 10. Mixed plastic. A plastic tub, bottle or pouch, or a plastic wrapper on each unit.
Certifications	10%	A product certification, meaning EPA Safer Choice, Cradle to Cradle, Green Seal or USDA Biobased, is worth 2 points and is not capped.Any other third-party mark is worth 1 point, and light marks count for at most 3 points in total.Score is points multiplied by 1.5, capped at 10. No marks listed scores 0.

Where the film cap applies. A dissolving polyvinyl alcohol film wrapped around each dose is per-unit plastic packaging, so it takes the packaging band down to the bottom one however good the outer box is: that is why Dropps and Cleancult sit at 3 out of 10 on packaging despite shipping in paperboard. A sheet whose dose is the polymer has no wrapper to cap, so it is scored for that polymer under the PVA criterion instead, and its outer packaging, usually a cardboard envelope, is scored on its own merits. That is how a sheet brand can take 9 out of 10 on packaging and still take the bottom band on PVA. Weights total 100 per cent. They are fixed before a category is researched and they are the same in every category, so a score in one table means the same thing as a score in another.

Not scored, and the pick rule

Where the published evidence cannot settle a criterion, the cell says Not scored and the criterion is left out of the total. The weights of the remaining criteria are rescaled, so a product missing one 20 per cent criterion is scored out of 80 rather than penalised with a zero. Ecover has no US price because it no longer runs a US shop, and that is a fact about geography, not about the product.

Rescaling has an obvious failure mode: it can only ever flatter a product that published less, because the criterion it dodged is the one it would have scored worst on. So a partly scored product is taken out of the ranking altogether. It keeps its number, it is labelled Partly scored wherever it appears, and it holds no position in the order and no place in a verdict, however high the rescaled figure is. Unranked on this build: Ecover ZERO All-in-One Tablets (cost per use); Finish Powerball Quantum Tabs (cost per use); Molly’s Suds Original Detergent Powder, Unscented (packaging); Tide PODS, Original (cost per use).

The rule is applied to this site’s owner on the same terms as everyone else. Where a Vove page did not state a pack size, we said so and left the product unranked until the figure was sourced.

What version 1 deliberately does not score

Cleaning performance. No laboratory testing has been run. There is no cleaning score, no grease score, no residue score and no drying score anywhere on this site, and there will not be one until a test exists. A performance number without a test behind it is a fabricated result, and publishing one would make every other number on the page worth less.

When testing does start we will publish the protocol before the results: the machine, the cycle, the soil load, the water hardness, the number of runs, and who scored them. The dishes will be photographed as they came out, including the ones that came out badly. Until that page exists, read these comparisons as what they are, a careful reading of the label.

Health outcomes. We do not say an ingredient causes or prevents disease. The polyvinyl alcohol question is presented with the study and the regulator side by side, because that is the honest state of it.

Review counts and ratings. Several brands here publish a review total that turns out to cover a whole product line rather than the pack on the page. We report those figures as line aggregates where we can see that they are, and we do not score them.

Corrections

If a figure here is wrong, tell us and we will fix it. Email hello@lowtoxreview.com with the page, the figure and where the correct one comes from. Corrections are made on the page itself and noted with the date of the change, and a correction that changes a score changes the verdict too. We do not quietly edit a number and leave the conclusion standing.

Brands are welcome to write in. A brand that publishes an ingredient list it did not publish before, or moves one from a help centre onto the product page, will score differently the next time the page is checked, and that is the intended effect.

How often this is re-checked

Prices move and product pages get rewritten, so every figure on this site carries the date it was read. When a page is updated, every price and every ingredient list on it is re-checked on the same day and the date at the top of the page changes with it. We do not refresh one cell and leave the rest of the table sitting at an older date.

Each live category is re-checked at least every three months, and sooner if a brand reformulates or a price moves enough to change a band. If you are reading a page whose date looks old, the figures are as of that date and nothing has been quietly refreshed behind them.