Every Dovelyne article is built in the same order. This page describes that order so you can judge the result rather than take it on trust.
We start from the decision, not the product
A guide begins with the choice a household is actually facing, written out plainly: what has to be decided, what the constraints usually are, and what tends to go wrong. Only after that do we look at what is on sale.
Working in this order matters. Starting from a list of products and writing backwards produces an article that recommends whatever was easiest to find, and reads like it.
Primary sources before anything else
For any claim about safety, money, health, schooling or legal entitlement, we go to the body that actually sets the rule: the regulator, the department, the standards body, the professional association, or the manufacturer’s own published documentation. We cite it, and we record the date we retrieved it.
Requirements differ by country, and often by state or region within a country. Where a rule is local we say which place it applies to rather than writing as though one country’s rule were universal.
We do not cite another blog as the source of a fact. We do not repeat a statistic we cannot trace to its origin. If a number is widely quoted online but no one can point to where it came from, we leave it out.
Product information comes from specifications
Product detail comes from the manufacturer’s specification, the official product page and the retailer listing, not from marketing copy and not from other people’s reviews rewritten.
Where independent hands-on testing exists and is credible, we can draw on it, and when we do we name it and link to it. That is a separate evidence level from testing something ourselves, and it is labelled differently on the page.
What we will not do
We do not invent testing, owner quotes, survey results, benchmark figures or expert comment. We do not describe using something we have not used. We do not present a manufacturer’s claim as an independent finding.
These are not aspirations. Checks in our publishing pipeline refuse content that contains first-person testing language without a dated test record behind it, and refuse a hands-on evidence label on the same basis.
When the evidence runs out
Research frequently ends without a clean answer. A specification is not published, a price cannot be confirmed, availability differs by country, or the honest finding is that two options are hard to separate.
When that happens we say so on the page, in a scope and limitations block, rather than filling the gap with a confident sentence. An article that tells you what it could not establish is more useful than one that hides it.
Related: how we verify, editorial policy and how we review products.