Content Freshness

This skill keeps content current enough to keep getting cited. Recency is a strong signal in AI answers — recently, substantively updated pages are cited markedly more, and some engines lean heavily on freshness because they retrieve in real time. But the signal rewards real updates: engines compare versions, so changing a date without changing the substance does nothing. This skill builds a cadence and a refresh method that earn the recency advantage honestly. It works for any business, B2B or B2C.

When to use this skill

• Setting an update cadence for a content library.
• Deciding which existing pages to refresh and how.
• Recovering pages that used to be cited and went quiet.
• Building freshness into an editorial process.

Before you start

Read aeo-foundation.md for the priority topics and prompt set — freshness effort should concentrate on the content tied to high-value prompts. If it is missing, point to the aeo-foundation skill.

Principles

• Substantive over cosmetic. A refresh must change real content — new data, new examples, revised guidance, added sections — not just the visible date. Engines detect the difference.
• Match cadence to volatility. Fast-moving topics (pricing, news, anything dated) need frequent updates; evergreen guides need less; reference/definition pages least.
• Prioritize by value. Refresh the pages that target high-value prompts and already have traction before touching the long tail.
• Recency is a tiebreaker, not a substitute. Freshness lifts content that is also well-structured, evidenced, and authoritative — it does not rescue thin pages.

Suggested cadence (adapt per topic)

Workflow

1. Inventory and flag. List priority pages with their last substantive update and target prompts. Flag anything stale relative to its type, and anything that lost AI visibility.
2. Diagnose decay. For pages that slipped, check what changed: outdated facts, a year in the title, superseded guidance, or a competitor publishing fresher work.
3. Refresh substantively. Update statistics and examples, add new sections or questions from the prompt set, correct anything outdated, and re-verify citations. Pull in evidence-and-citations for new data and answer-engineering for any restructuring.
4. Update the date honestly. Set dateModified (and any visible "updated" line) to reflect the real change — see the schema cookbook.
5. Signal the update. Ensure the page is re-crawlable (sitemap, internal links) so engines pick up the new version.
6. Schedule the next pass. Assign each page a cadence so freshness is systematic, not reactive.

Output format

• A refresh plan: which pages, what to change, target cadence, and owner.
• For each refreshed page, a note on the substantive changes made.
• Flags for pages that should be consolidated or retired rather than refreshed.
• (assumption) tags where update history was unknown.

Common pitfalls

• Cosmetic date bumps with no real change — detected and ineffective.
• Refreshing low-value pages while flagship content goes stale.
• Treating freshness as a fix for thin or unstructured content.
• No schedule, so freshness happens only when someone remembers.

Related Skills

• answer-engineering — restructure sections during a refresh.
• evidence-and-citations — swap in current data and live sources.
• ai-visibility-tracking — surfaces pages losing citations, triggering refreshes.
• structured-data — keep dateModified accurate and honest.