Measurement and Content Performance

Apply ../00-blog-writing-guardrails/SKILL.md first; its originality, accuracy, and pre-publish gates apply to every claim and every refreshed passage this skill produces.

1. KPI-to-funnel-stage gate

• Map each blog KPI to the funnel stage it actually measures before reporting it as a success signal: impressions and average ranking position measure discovery; organic clicks and click-through rate (CTR) measure the search-result decision; and time-on-page, scroll depth, and return visits measure on-page engagement — do not report a discovery-stage metric (e.g., impressions) as proof of an engagement-stage outcome. [BACKLINKO-CTR-01]
• Treat CTR as position-dependent, not a flat conversion target: average organic CTR for the #1 ranking position was 27.6% versus a small fraction of that by position #10 in a first-party analysis of over 4 million Google search results, so compare a page's CTR only against pages holding a similar ranking position, not against a single blanket benchmark. [BACKLINKO-CTR-01]
• Because the top 3 ranking positions captured the majority of clicks in the same analysis, prioritize headline and snippet testing on pages already ranking in or near positions 1-3, where a CTR gain converts into the most absolute clicks; a CTR improvement on a page ranking below position 10 moves little absolute traffic. [BACKLINKO-CTR-01]
• Report engagement metrics (time-on-page, scroll depth) and conversion metrics (email signups, product clicks) as a separate KPI layer from traffic metrics; a page can gain impressions and clicks while engagement or conversion falls, and that split must be visible in the same report, not collapsed into one topline number. [BACKLINKO-CTR-01]

2. Content-decay detection and refresh-cadence gate

• Flag a published post for review when its organic clicks or ranking position show a sustained decline (not a single-week dip) relative to its own prior baseline, rather than reacting to normal week-to-week search-visibility noise. [GOOGLE-EEAT-01]
• Distinguish decay caused by stale, outdated, or superseded information from decay caused by external ranking-factor or competitive changes before choosing a fix: a post whose facts, prices, screenshots, or examples are outdated needs a content refresh; a post that is still accurate but has been outranked needs stronger originality, depth, or first-hand evidence, not just a datestamp change. [GOOGLE-EEAT-01]
• When refreshing a decayed post, materially update the substance — corrected facts, new data, added first-hand detail — and update the visible and structured-data modification date to match; do not change only the datestamp without a substantive content change, since freshness signals must reflect real, verifiable updates. [GOOGLE-EEAT-01][GOOGLE-SCHEMA-01]
• Re-verify every statistic and claim carried into a refreshed post against its current primary source before republishing, exactly as required for a new post; a refresh does not exempt a claim from the accuracy gate. [GOOGLE-EEAT-01]