Credible Roots / Corrections
Corrections
Corrections
A site that tells people to distrust unsourced claims should be checkable itself. When something here is wrong, it gets fixed on the page and recorded here, because a silent edit is indistinguishable from having been right all along.
How corrections are handled
- Factual errors are corrected on the page and logged here, with the date and what changed.
- Substantive rewrites get a new dateModified in the page's structured data. Typo fixes and formatting changes do not, because refreshing a date without changing the substance is exactly the practice this site argues against.
- Disputed readings — where a guideline is genuinely ambiguous rather than misstated — are noted in the piece itself rather than resolved silently in our favor.
Found something wrong? Tell us. A specific quote and a source is the fastest route to a fix.
Correction log
14 August 2026. The personal branding page for doctors was withdrawn shortly after publication. Health is a Your Money or Your Life topic, where Google weights demonstrated expertise most heavily and where being slightly wrong about HIPAA or state board advertising rules would carry real consequences for a reader. We are not the right authors for that page. The audience page for founders and the one for real estate agents remain.
14 August 2026. The personal branding pages were reorganized into a hub with three audience pages at founders, doctors and real estate agents. Two overlapping pages — one for professionals with 15+ years, one on founder publishing — were merged into the hub and the founders page respectively, and both old URLs redirect. We considered and declined pages for audiences we have not worked with; writing them would have been a traffic decision rather than an expertise one.
14 August 2026. Eighteen pages were retired and merged into four. The five profession-specific Wikipedia pages, the two overlapping notability notes, the four Wikidata notes and the Wikidata service page, the two knowledge-panel pages, and three positioning pages each covered ground that read better as one reference than as several. Nothing factual changed; the retired URLs redirect permanently to Wikipedia notability, Wikidata, Google knowledge panels and how this ends up at Wikipedia. We did this in the site’s first week deliberately, while nothing depended on those URLs.
14 August 2026. Two tables previously carried a figure of roughly 60% for the reach penalty on LinkedIn posts containing an external link, attributed to “benchmark studies”. We could not trace it to a named published source, so the number has been removed from both pages and the effect is now described as a practitioner report. Affected pages: Doing it yourself and What founders should actually post on LinkedIn.
The site published on 13–14 August 2026. Entries appear here as corrections are made.
Claims we decline to repeat
Separate from our own corrections: these are figures and rules repeated widely in this industry that do not survive a check. Where a piece here touches one, it says so rather than passing it along.
| The claim | The problem | Where we address it |
|---|---|---|
| The 7-11-4 rule — seven hours, eleven touch points, four platforms before trust, attributed to Google | It does not appear in any Google publication, and practitioners who have gone looking have said openly that they cannot find a source. Google's 2011 Zero Moment of Truth research is real; these numbers are not in it. | Touch points before trust |
| llms.txt as a requirement for AI search visibility | It is a proposal, not an adopted standard, and no major answer engine requires it. Sites are sold it as though it were table stakes. | Does llms.txt actually do anything |
| Guaranteed Wikipedia articles, and knowledge panels available for a fee | Notability is decided by volunteer editors and no vendor controls the outcome. Knowledge panels have no submission process and no paid route. | What a Wikipedia page actually costs |
| A ~60% reach penalty for LinkedIn posts containing an external link | Repeated everywhere as a benchmark figure. We could not trace it to a published study, so we no longer print the number; we describe the effect as reported by practitioners instead. | Doing it yourself, LinkedIn for founders |
If you have a source for any of the above that we have missed, we would genuinely like to see it, and we will correct this page and say so.
Related
- Editorial standards — how a field note is researched, drafted and reviewed.
- About the author — who stands behind the bylines.