Network editorial standards

1. Intent before keywords

Every brief starts with the search intent, not the keyword volume. If the query is answered best by a calculator, we build a calculator. If it is answered by a comparison table, the table is the page — not an 1,800-word warm-up before it. Pages that exist only to occupy a keyword are rejected at brief stage.

2. Information gain is mandatory

A page must add at least one of: original calculation or model, first-hand testing, a practitioner interview, a primary-source reading that competitors got wrong, a reusable template, or a dataset. Rewriting the current top ten results is not a brief.

3. Sourcing hierarchy

1. Statute, regulator publications, official government data and standards bodies.
2. Provider documentation and filed policy wordings, cited with a date.
3. Peer-reviewed or institutional research.
4. Named practitioner interviews.
5. Reputable trade press, only for context — never as the sole support for a factual claim.

Statistics are attributed to whoever produced them. If we cannot reach the primary source, we do not publish the number.

4. Expert review for YMYL

Insurance, legal, tax, credit, health and safety content is reviewed by a named practitioner before publication and re-checked at each substantive update. Reviewers never review their own writing, and their role is stated on the page.

5. Freshness

Pricing and comparison pages: reviewed at least quarterly. Regulation-linked pages: reviewed when the rule changes. Everything else: annually. The visible Updated date changes only when content changes.

6. Link policy

• Outbound citations are chosen for the reader; most earn us nothing.
• Affiliate and lead-gen links are disclosed on the page where they appear.
• We do not buy links, sell links that pass ranking signals, or run reciprocal link agreements.
• Cross-network links are contextual, per-article, and never templated across a whole site.
• Anchors are descriptive and natural, not stuffed exact-match phrases.

7. AI policy

Assistive use is allowed for research organisation, outlining and editing. Prohibited: publishing unedited generated text, fabricating quotes, fabricating credentials, generating volume for its own sake, or scraping and respinning competitor content.

8. Advertising and user experience

Ad slots are reserved at fixed sizes so they cannot shift the page while it loads. No interstitials over article content, no auto-playing audio, no ads inside comparison tables. Core Web Vitals targets for every template: LCP under 2.5s, INP under 200ms, CLS under 0.1, per web.dev's Core Web Vitals definitions.

9. Corrections and retractions

Factual errors are corrected with a dated note. Serious errors get a retraction notice at the top of the page. Pages are not silently deleted; retired URLs redirect to the closest useful replacement.

10. What gets a page killed

• Unsourceable claims.
• Guaranteed outcomes ("approved in 24 hours", "cut your premium in half").
• Undisclosed commercial relationships.
• Doorway pages: near-identical location or variant pages with no unique local substance.
• Reviews of things nobody evaluated.

Reference documents we hold ourselves to

• Google Search Essentials
• Creating helpful, reliable, people-first content
• Google spam policies, including link spam
• Structured data guidelines
• Search Quality Rater Guidelines (E-E-A-T)