Method
Sources and Methodology
We build each guide from primary documentation, test what can be tested, label what is illustrative, and date everything. This page explains the hierarchy of sources and the checks every page passes.
Source hierarchy
| Tier | Examples | How we use it |
|---|---|---|
| 1. Primary documentation | Google crawling documentation, Google Search Central, Search Console Help, Bing Webmaster documentation | Basis for every factual claim about search engine behavior |
| 2. Standards | RFC 9309 (Robots Exclusion Protocol), sitemaps.org protocol, ISO language and region codes | Definitions and limits |
| 3. Official statements | Search Central Blog, documentation changelogs | Dates of changes; quoted only when the post text was read |
| 4. Regulators and courts | Data protection authorities, court judgments | Privacy context, such as log retention |
| 5. Our own tests | Rendering comparisons, robots.txt unit tests | Only for things we can reproduce, with method stated |
We do not cite SEO tool vendors' studies, unnamed "industry data" or social media posts as evidence for how Google works.
Checks before publication
- Each quote is matched against the current version of its source page, and the page's "last updated" date is recorded.
- Each external link is opened and confirmed to support the sentence it is attached to.
- Numbers (limits, thresholds, quotas) are checked twice against the source.
- Constructed examples use
.examplehosts and carry an "Example (constructed)" label. - Before release, a separate fact-check pass (AI-assisted, then confirmed by the editor) compares every claim with its source.
How our tools are tested
The robots.txt rule explorer runs a set of unit tests built from the examples in Google's robots.txt documentation, covering group selection, merging, longest-match precedence, ties, wildcards, end anchors, case sensitivity and percent-encoding. The tests are kept with the site's build source and are available on request. The audit checklist performs no analysis; it records what a person ticks.
Known limits
- We describe documented behavior. Search engines can behave differently from their documentation, and individual sites can see results we cannot predict.
- Our guides focus on Google. Other search engines and AI crawlers may not render JavaScript or follow the same rules.
- We are not lawyers. Where a guide touches privacy or licensing, it points to the official source and suggests checking with your own advisers.