What we will and will not do
Independence
We do not run paid placements. A vendor cannot pay to be added to the directory, to be moved up within a category, or to soften a verdict. We do not accept gifts, paid trips, or equity from vendors we cover. Some pages contain affiliate links – they are disclosed on the page they appear on, and they do not influence inclusion or assessment. If we ever decide a particular vendor relationship is incompatible with neutral coverage, we end the relationship rather than soften the coverage.
Corrections policy
If we get something wrong, we say so on the page and timestamp the change. Material corrections – a wrong price, a wrong feature claim, a wrong attribution – appear in a short note at the end of the page so a reader can see the prior version. Minor edits (typos, broken links, cosmetic rewording) are made silently. To request a correction, email hello@serp.fast with the page, the specific claim, and a verifiable source. We respond within five business days.
Affiliate and sponsorship disclosure
Outbound affiliate links are marked with rel="noopener nofollow sponsored" and a visible disclosure on any page that contains one. Newsletter placements and news-feed entries that are sponsored are labelled as sponsored on the surface they appear on. We never re-publish a vendor's copy as if it were our own; sponsored copy is written by the sponsor and presented as such. Affiliate revenue does not buy access to editorial pages, including the methodology we use to evaluate tools.
Source attribution
When we cite a number we cite the source – vendor documentation, a third-party benchmark, or our own test harness described on the how-we-test page. When we paraphrase another publication's reporting we link to it. Quotes are quoted; claims that came from a conversation with a vendor are attributed to that conversation and dated. We do not present vendor-supplied charts as our own measurement.
AI-assistance policy
Some of our drafts are produced with AI assistance. A human editor reviews and rewrites every published article. We do not publish machine-generated copy as if it were written by a person. Where an AI system was used to gather, summarise, or translate source material, the article says so. Test data, scoring, and verdicts are produced by people – not by an LLM grading another LLM.