Editorial policy
These rules apply to every comparison on ProductPick. They set out how we choose products, which sources we use, how the score is calculated and how we correct mistakes.
Scores and rankings are calculated from the criteria.
Commission, advertising and partnerships never affect the score or the ranking, and nobody can buy a better place in a comparison.
Independence
The editorial team decides which product categories and models to compare and how to assess them. Manufacturers and retailers cannot see or approve a comparison before it is published, and they cannot buy a score, a place or a position among the models compared.
Selection
We choose models on sale in the UK that cover different price points and needs. The selection is based on relevance and quality. We work with some retailers through affiliate networks such as Addrevenue, and may work with other networks in future. Where two models are of comparable quality, we give preference to the one available from such a retailer. Well-known reference models are included even when we earn nothing from them. Products that are no longer sold, or are not in stock at any UK retailer, are removed.
Source requirements
We use three kinds of evidence and keep them apart:
- Manufacturer information: product page, manual and data sheet.
- Independent tests by consumer organisations and testing publications, such as Which?, where they exist for the model. We never reproduce results that sit behind a paywall, only what is freely visible. Where no such tests exist we use review sites, and only say a model was tested when the site measured it itself.
- Customer reviews from retailers and price comparison sites with at least 10 reviews. We give more weight to the number of reviews, how recent they are and verified purchases than to the star average.
We only use sources we have actually opened. Every detail shows what kind of source it comes from, and every page ends with links to the evidence. If a detail cannot be backed up, we write not stated.
How the score is calculated
- Criteria. Every product category has 5–7 criteria with weightings that add up to 100 per cent. The weightings are shown on every comparison.
- Points. Each model gets 0–100 points per criterion based on the sources.
- Score. The score is the weighted average divided by 10. It is calculated automatically and never typed in by hand.
- Ranking. The order of the comparison follows the score. The model with the highest score is our top pick.
The bar next to each product shows how much each criterion contributes to the score. The price assessment is always based on the lowest price we found, not on the price at the retailer we link to.
Awards such as “Best value” or “Best for small kitchens” are given by use case, so that you can find the right model even if our top pick does not suit you. They are based on our comparison of the sources and never on a test of our own.
Language and claims
- We never write that we have tested, tried or handled a product. We report what the manufacturer, the retailer or customer reviews say.
- “Our top pick” and “best” describe the result of our comparison of independent testers' findings, customer reviews and specifications. They never refer to a test of our own. We only write “tested” or “test winner” when a named independent tester has tested or awarded the model.
- We make no health claims.
- We do not quote sources at length, at most a short phrase, and we write in our own words.
Commission and advertising
Some buy buttons are affiliate links and are labelled Affiliate link next to the button. If a product is in stock at a retailer we work with, we link there, even if that retailer is not always the cheapest. If the product is cheaper elsewhere, we show the lowest price next to the button. A page may have a paid placement at the top, always labelled Ad. It shows the product's real score and rank and never changes the score or the ranking. Read more on our advertising page.
AI in our work
We use AI to gather and organise the research and to write the copy. The criteria models and every comparison are approved by a member of the editorial team before publication, and every page is checked separately against its sources.
Updates
Every comparison is reviewed at least every three months, and sooner if a new model is launched, the price changes significantly or a new independent test is published. The date of the latest assessment is shown at the top of every page. If our top pick is no longer sold, we update the comparison.
Corrections
If we or you find a mistake, we check the detail against the sources. If you are right, we correct the page, recalculate the score if the points change, and update the date of the latest assessment. You can report mistakes through the contact page.
Read more about how the editorial team works on the editorial team page.