AEO / GEO

What this channel does

Gets Redis cited and recommended by answer engines instead of summarized away. We can't rank our way out (most traffic already sits at positions 1–3) — the play is to become the source the answer quotes, and to convert the visits that still land.

The play

• Machine-legible by construction — schema, structured sections, tables, and bottom-line-up-front blocks on every new page. New components in the design system get designed around what AI engines consume well.
• Glossary recovery — verdict pass on every page (keep / fix / retire), expansion list for missing terms, depth an answer box can't replicate; images and video on kept pages.
• Blog topic hubs — tag-driven hubs targeting commercial LLM terms (context engineering, LLM cost, RAG + vector), cross-promoting accumulated posts to earn the head terms.
• Head-term discipline — every new page owns one major head term from the keyword portfolio; no cannibalization, visible gaps.
• Human-authored authority — the multi-team publishing model replaces AI-volume content: our people write, technical depth, named series. Thin AI content is a losing strategy long term.

Standing rules

• The naming gate — answer engines repeat whatever we publish; product and capability names clear PMM before pages ship.
• Dev marketing owns "agent recommendation" from the advocacy/content side (assigned Sep 1) with a close partnership into this program; docs stays product-owned (Michelle partnership).

Measured by

Citation share on Profound against a locked baseline, organic conversion rate on treated pages, and recovered coverage on the head-term map. Traffic alone is not the scoreboard — visibility without visits is the problem we're correcting for.

Current state

Erosion quantified (~854K projected annual click loss, base case); glossary and blog verdict passes queue behind the keyword-to-content map; the incoming agency carries the audit legwork.