Loading...
The playbook Moonrank's SEO Agent uses to turn a topic into a plan a writer can start on Monday — with the order set by what is beatable, not by what is biggest.
Download the methodology, see the winnability formula, or run it live inside the SEO Agent.
The Content Plan Builder skill is how Moonrank's SEO Agent goes from a seed to a sequenced plan. It expands a topic, a keyword list or the site's own rankings into a keyword universe with real metrics, clusters it into topics a writer would recognise, and scores every cluster on demand weighted by how beatable its difficulty is.
The order that produces is the point. A cluster with a fifth of the demand and a quarter of the difficulty comes first, because it is the one that will rank. The roadmap then puts each cluster's pillar before its supporting pages, gives every article a linking note, and ends by proposing the first phase straight onto the content calendar for the user to apply.
It's written as a SKILL.md, a plain-language instruction format that people and AI agents can both follow. Read it to plan your own quarter, hand it to a content team as the brief, or let the SEO Agent build it from a seed.
“Build me a content plan around cold brew coffee.”
287 keywords in six clusters. Two are quick wins — 'how to make cold brew at home' at 41,000 searches and a mean difficulty of 16 leads the plan at a score of 92. 'Best cold brew makers' is the biggest cluster by volume and lands third: strategic, difficulty 44, attacked in Phase 2 once the guide is ranking and can pass it authority. Phase 1 is the guide plus three supporting pieces, each linking up to it; Phase 1's twelve keywords are proposed onto your calendar for you to apply.
A topic, a keyword list, or the site itself — and a universe size if you want more or less than 300.
Expanded, deduplicated, and looked up in batches — real volume, difficulty and CPC, never estimated.
Six to fifteen topics a writer would name, each scored on demand weighted by winnability.
Phases with pillars before supporting pages, and Phase 1 offered onto the calendar.
Every keyword looked up; a keyword with no metrics is marked, never guessed.
Split by intent, grouped by theme, one page per cluster.
Log-normalised demand times winnability, so one giant cluster cannot swamp the rest.
Quick win, Strategic, Fill — the same vocabulary on every plan.
Pillar before supporting, quick wins before strategic, each item with its rationale.
Up to the pillar, down to the supporting pages, across to one sibling cluster.