Research
We start from what can be verified: the vendor’s own documentation, pricing pages, trial terms and security attestations, plus our shared facts file. For articles we add primary sources such as regulator guidance where they help the reader, and we link to them.
Writing
Lists, profiles and comparisons are written in an evaluator’s voice. We describe what the documentation and published facts show, and we turn things only a trial or demo can show into questions and checklists for the reader. We do not write “we tested” unless we did, and we do not quote star ratings from review platforms.
Each provider claim must trace back to a fact: a feature on the checklist, a certificate, a price, a founding year, a head office or the vendor’s own stated focus. If we cannot trace it, we cut it or mark it as a question to ask.
Fact check
A second editor checks every number, feature claim and price before a page goes live, and re-runs the scoring if a fact changed. Scores on the site are computed at build time from the facts file, so a corrected fact flows into every list, profile and comparison together.
Monthly re-check
Prices, trials, certifications and features are re-checked every month. Pages show the month of the last check. Articles carry their publication date and, when changed, an updated date.
Corrections
If you find an error, tell us. We correct factual errors quickly and note material corrections on the page. Vendors may send updated facts with a public source; we do not accept changes we cannot verify.
Partners
Some links are partner links, disclosed in the footer of every page. Partners do not see pages before publication, cannot request changes to scores or wording, and are scored with the same code as everyone else. See the advertiser disclosure.
Use of AI tools
We use software, including AI tools, to help draft and check text, run the scoring and catch inconsistencies. Editors review and take responsibility for everything published.