How we score, and how we choose

Standing in a shop with two boxes of dishwasher tablets, you cannot run a test. What you can do is read. It turns out that reading carefully separates the field more than most people expect. Of the 87 products we have scored so far, eleven contain polyvinyl alcohol, and six of those do not say so on the product page; twelve are made with shedding plastics, and eight of those do not say so on the product page. So we score what a reader can check: whether the brand publishes its ingredient list, what is on that list, whether the product puts a dissolving plastic down the drain, what one use costs, what the packaging is made of, and which third-party marks it actually carries. Every one of those is a fact on a page, and every page is linked and dated.

One more thing, said plainly because it matters. This site is owned by Vove, and Vove makes some of the products we score. The rules below were written so that they could be checked rather than trusted: the weights were fixed before any category was researched, the bands are printed, and our own products go through exactly the same arithmetic. The scores are the record; the order of each page is then our editorial choice, made in the open. It often follows the scores and it does not have to: we front the products we make and would buy, we give newer and smaller brands the benefit of the doubt where the published record is thin, and every table says beside it that the order is ours. Where our own products lose a criterion or a score, the table says that too.

Each product gets a score from 0 to 10 on every criterion, the scores are weighted as in the table below, and the total is the number you see on the card. The weights add up to 100 and they are the same in every category. Cost per use is the one criterion scored relative to the table: the cheapest use on the page scores 10, the dearest scores 2, and the rest sit evenly in between by rank. The exact rule is in the table.

What the colours mean. A score of 7.0 or above is shown in green, 5.0 to 6.9 in amber, and anything below 5.0 in red. The colour is only the number, said twice.

This is the list every published ingredient list is read against. It was fixed before the read, it is the same for every brand including ours, and it has two tiers. Anything from the upper tier in a product puts that product in the third band on the ingredients criterion. Anything from the lower tier alone puts it in the second. A list with nothing from either tier takes the top band. A product that publishes no list takes the bottom band, because we cannot clear what we cannot read.

Two things the list is not. It is not a claim that a product with a lower-tier entry is unsafe: a silicone antifoam in a dishwasher tablet or a disclosed essential oil in a hand soap is something a careful household may want to know about, not a hazard at that dose. And it is not a medical document. We do not say an ingredient causes or prevents anything. Each entry carries the one-line reason it is on the list, and if you think an entry is wrong, or missing, tell us and we will say so on this page.

Where the fragrance line is drawn. A fragrance counts as disclosed when the page names the substances in it, whether as essential oils or as INCI allergen names such as limonene or linalool. A scent name on its own, Lime and Eucalyptus or Lemongrass, names a smell and not an ingredient, and counts as undisclosed. The same line is applied to our own products: our toilet tablet failed it until 26/08/2026, when its page named the components behind the scent and the read was repeated, and our hand soap passes it.

On the current read, seventeen of the 87 products carry something from the upper tier of the watch list, and every one of them publishes a list.

Sometimes a page simply does not give us what we need: no price, or no pack size, so there is no cost per use to work out. In that case the cell says Not scored, that criterion is left out, and the remaining weights are scaled up so the product is scored out of what we could read rather than punished with a zero.

But a product scored on fewer things is not comparable with one scored on all of them, and the missing criterion is usually the one it would have done worst on. So a partly scored product keeps its number, is labelled as partly scored wherever it appears, and is not ranked and never gets a verdict. Right now that means: 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); Vove Dish Soap Bars (cost per use, packaging); Blueland Dish Soap Powder Starter Set (cost per use); Puracy Dish Soap, Green Tea and Lime (cost per use, packaging); No Tox Life DISH BLOCK Solid Dish Soap (cost per use, packaging); Dawn Ultra Dish Soap, Clean Scent (cost per use); Puracy Plant Powered Gel Hand Soap (packaging); Full Circle 100% Plastic-Free Loofah Scrubber Sponge (cost per use, packaging); Full Circle Walnut Shell Scrubber Sponge (packaging); Marley’s Monsters Washable Sponge, Juicy Fruits (packaging); Scrub Daddy Scrub Daddy Original (packaging); Scotch-Brite Heavy Duty Scrub Sponges (cost per use, packaging); Full Circle The Daily Dish Brush (packaging); OXO Dish Brush with Scraper (packaging); Full Circle The Upstanding Bottle Brush (packaging); OXO Bottle Brush (packaging); Smart Sheep Original Wool Dryer Balls, 6-Pack (packaging); Tru Earth Reusable Wool Dryer Balls, 4-Pack (packaging); Bounce Outdoor Fresh Dryer Sheets (cost per use, packaging); Stasher Sandwich Bag (packaging); Bee’s Wrap Assorted 3 Pack (packaging); Etee Beeswax Food Wraps (packaging); (re)zip Reusable Sandwich Bag, Flat (packaging); Ziploc Storage Bags, Quart (cost per use, packaging); Molly’s Suds Oxygen Whitener (cost per use, packaging); Puracy Laundry Stain Remover (cost per use); OxiClean Versatile Stain Remover Powder (cost per use, packaging); Clorox Ultra Clean Toilet Tablets Bleach and Blue (cost per use); The Konjac Sponge Company 100% Pure Konjac Big Buffer Sponge (packaging); Bathorium Konjac Walnut Shell Exfoliating Body Sponge (packaging); Boie Body Scrubber (packaging); myHomeBody Konjac Bath Sponges for Babies and Kids (packaging); Bathorium Konjac little charlie Body Sponge (packaging); Frida Baby Replacement Sponges for the 3-Step Cradle Cap System (packaging); myHomeBody Natural Konjac Facial Sponges (packaging); Boie Face Scrubber (packaging); Dew Puff Original Konjac Sponge (packaging). That rule has applied to our own products too: where a Vove page did not state a pack size, we said so and left the product unranked until it did.

Who gets recommended. The products that took a verdict on a page, plus the highest scoring products until at least three carry the label. Where the table is short on fully scored products, a partly scored one can be promoted, and its label says so.

Cleaning performance. No laboratory test has been run, so there is no cleaning score, no grease score and no drying score anywhere on this site. A performance number without a test behind it would be an invented number, and it would make every real number on the page worth less. When testing starts we will publish the protocol before the results: the machine, the cycle, the soil, the water hardness, the number of runs, and who scored them, with photographs of the dishes as they came out, good and bad.

Health outcomes. We do not say an ingredient causes or prevents a disease. On polyvinyl alcohol, for example, we put the study and the regulator side by side, because that is the honest state of the evidence.

Review counts. Several brands print a review total that covers a whole product line rather than the pack on the page. We say so where we can see it, and we do not score it.

If a figure here is wrong, tell us at [email protected] with the page, the figure, and where the right one comes from. Corrections are made on the page and dated, and a correction that moves a score moves the verdict with it. We do not quietly fix a number and leave the conclusion standing.

Brands are welcome to write in. A brand that starts publishing its ingredient list, or moves it from a help centre onto the product page, will score differently the next time we read the page. That is the intended effect.

Every price is the US dollar figure we read on the day shown, and that is the only figure cost per use is calculated from. Where you see a pound or euro amount beside it, that is a rough conversion for orientation and is scored on nothing. When a page is updated, every price and every ingredient list on it is re-read the same day and the date at the top changes. We do not refresh one cell and leave the rest at an older date.

Each 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 the date on a page looks old, the figures are as of that date and nothing has been quietly refreshed behind them.