Editorial Policy & AI Use

How we research

• We start from a specific runner problem (arch collapse, overpronation, wide forefoot, knee pain) rather than from a product list.
• Sources include manufacturer specifications, retailer listings, published lab and wear-test reporting, podiatry and sports-medicine literature, and aggregated runner feedback.
• Claims about stability technology, drop, stack height, weight, and width options are checked against the manufacturer's own spec sheet.
• Where evidence is thin or contested, we say so instead of inventing certainty.

AI assistance and human review

• We use AI tools to help with research synthesis, outlining, and first drafts.
• No article publishes automatically. Every draft is reviewed and approved by a human editor before it goes live.
• Review checks factual accuracy, product specs, originality, and that the article adds something a reader cannot get from the top existing results.
• Drafts that are thin, duplicative, or unverifiable are rejected rather than rewritten into filler.

This reflects Google Search's guidance: how content is produced matters less than whether it is helpful, original, and people-first — and automation used purely to manipulate rankings is spam.

Ratings

Scores weigh arch and medial support, midsole stability, fit and width availability, durability, and value. A score is never adjusted for commercial reasons, and products we would not recommend are simply left out or explicitly flagged.

Updates and corrections

• Roundups are reviewed on a rolling basis and updated when a model is replaced, discontinued, or significantly changed.
• Substantive corrections are made in place and noted where they materially change a recommendation.
• Spotted an error? Email arendwier@gmail.com and we will verify and fix it.

Independence

Advertising and affiliate relationships never influence rankings, scores, or which products we cover — see the Affiliate & Advertising Disclosure. Medical scope and limitations are described in our Terms of Use.