Skip to content
All topics12 topics
Page

Editorial standards and research method

How Light Scripts researches buying guides: primary sources, verified quotes, how we recommend, how we use AI, and how we handle corrections.

Light Scripts publishes buying guides for home-office equipment. This page explains how we research, how we decide what to recommend, how we handle sources and corrections, and where the limits of our work are. It is the standard we hold every article to.

Who is responsible

Every article is produced by the Light Scripts editorial team and reviewed before it is published. If you find an error, tell us through the contact page and we will correct it. We are an independent publication, and our editorial choices are never decided by commercial relationships.

How we research

We gather evidence before we write.

  • We search the live web and fetch the pages we rely on, rather than working from memory.
  • We prefer primary sources: standards bodies, government and regulator guidance, manufacturer documentation, and independent testing.
  • When a claim depends on a source, the writer must quote it, and our system checks that the quote actually appears in the page we fetched. Notes that cannot be verified against the fetched text are not used.
  • We record the source URL for every note so a reader or an editor can check it.
  • When the evidence for an angle is thin, we say so rather than fill the gap with assumptions.

How we choose what to cover

We write about the decisions buyers actually face, organized around the reader's problem rather than a list of products. We favour fewer, well-justified recommendations over padding an article with alternatives. A product is included when the evidence supports it and left out when it does not.

How we recommend

  • We separate manufacturer claims from independently reported findings from our own analysis, and we label which is which.
  • We explain who a product suits and who should avoid it, and we make the trade-offs explicit.
  • We do not call a product the best overall unless the evidence supports that, and we explain the conditions under which each recommendation makes sense.
  • We do not invent specifications, measurements, prices, discounts, availability, warranty terms, ratings, review counts, or test results. When a number cannot be attributed to a source, we describe the relationship instead of stating the number.

What we do not claim

Unless an article says otherwise, we have not physically tested the products we write about. We do not claim hands-on testing, personal use, or measurements we did not perform. When we describe performance, we attribute it to the source that reported it.

AI assistance

Some drafts are prepared with AI assistance. That assistance is used to draft and to organize research, not to invent facts. Every draft is built from sources our system fetched and verified, and it is reviewed before publication. If a claim cannot be supported by a source, it is removed. We disclose this because you deserve to know how an article was made.

Sources and citations

Articles list the sources they relied on. We cite authoritative material by name where a claim depends on it, and we never invent a study, a test result, or a source we are not certain exists. Where a guide would benefit from a source we could not verify, we prefer to say less rather than guess.

Updates and corrections

  • We revise articles when the underlying products, prices, or standards change.
  • When we correct a factual error, we fix it rather than quietly delete it, and the updated date on the article reflects the change.
  • If you spot a mistake, email us. Corrections from readers are welcome and taken seriously.

Contact

For corrections, editorial questions, press, or partnership enquiries, use the contact page.