Building SEO Pages from Search and User Evidence
Ground each SEO draft in live SERPs, linked user quotes and verified product facts
Definition
An evidence-grounded content workflow uses current search data to decide what a page should cover, primary-source user language to supply non-generic material and a human review gate before publication.
Perspectives
Charlotte Schmitt (2026-09-07, X)
Scope one new or refreshed page, check for cannibalization, and let keyword volume, difficulty and a live SERP determine the format, length and FAQ topics. Use buyer-intent formats such as competitor alternatives, best-of lists, comparisons and pricing guides rather than asking a model to invent a generic article from a thin prompt.
Mine 6–10 verbatim, source-linked quotes from Reddit, Hacker News, X or another primary-source corpus, and make the dominant pain the page's hook. Verify prices and product claims against official sources at writing time; represent competitors with honest pros and cons and disclose the vendor relationship.
Integrate the draft with internal links, redirects, sitemap, assets, metadata and appropriate schema, then run link, consistency and honesty checks. Hand off a draft or pull request for a human publication decision instead of mass-generating, reusing sibling-page copy or auto-publishing.
How to apply
- Fits a site that can access current SERP or keyword data, collect attributable customer language and budget human review time; without evidence, stop and ask rather than filling gaps from model memory.
- Choose the page class with SaaS SEO Page Type Priority, then deepen the query brief with Building GEO Content Briefs from Search Surfaces.
- Use Pruning SEO Pages by Search Performance for accumulated thin pages instead of treating this production workflow as a cleanup rule.
- Keep publication behind a human approval boundary and evaluate the live page with dated SEO baselines.
Limits
- The workflow and ranking claims come from Octolens's founder and its own open-source skill description; no controlled comparison isolates search data, user quotes, editing or pruning.
- API pricing, coverage, quote freshness, source-link durability and the reviewer time per page are not reported.
- Verbatim social quotes can require privacy, context and reuse-rights review even when the source link is preserved.
- Search results, prices, competitors and schema requirements change, so evidence captured for one draft expires.