Develop Data-Driven Content Strategies

Plans content that drives traffic, builds authority, and generates leads by being searchable, shareable, or both - prioritizing search traffic as the foundation. Before planning, it checks for an existing .agents/product-marketing-context.md file and only asks about business context, customer research (pre-buying questions, sales objections, support-ticket themes), current content state, and competitive landscape not already covered there. For searchable content it targets a specific keyword or question, matches search intent exactly, and structures headings to mirror search patterns; for shareable content it leads with a novel insight or counterintuitive take and tells stories that make people feel something.

It identifies the 3-5 core pillars a brand should own through four lenses - product-led (what problems the product solves), audience-led (what the ideal customer needs to learn), search-led (what topics have volume), and competitor-led (what rivals rank for) - checking each candidate against pillar criteria: alignment with the product, fit with what the audience cares about, real search or social interest, and enough breadth to spawn many subtopics.

It catalogs concrete content types: use-case content (a persona-plus-use-case formula for long-tail keywords), hub-and-spoke structures (a comprehensive hub with interlinked subtopic spokes, reserved for major topics with real depth rather than every blog post), template libraries, thought leadership, data-driven content (product data, public data, or original research), expert roundups of 15-30 contributors, case studies structured as challenge-solution-results-learnings, and "meta content" behind-the-scenes transparency pieces. It maps keyword modifiers to buyer stage - "what is/how to/guide to" for awareness, "best/top/vs/alternatives" for consideration, "pricing/reviews/demo/trial" for decision, "templates/tutorial/setup" for implementation - and mines six ideation sources: keyword exports (Ahrefs, SEMrush, GSC), sales or customer call transcripts, survey responses, forum research (Reddit, Quora, Indie Hackers, Hacker News), competitor content analysis, and direct sales/support input. Content ideas are then scored on a weighted rubric: customer impact (40%), content-market fit (30%), search potential (20%), and resource requirements (10%).

Use it when deciding what content to create, in what order, and for which audience, building topic clusters or an editorial roadmap, or when the need is strategy and prioritization rather than copywriting itself. It delegates out: copywriting for individual pieces, seo-audit for technical/on-page SEO, ai-seo for AI-search optimization, programmatic-seo for content at scale, site-architecture for URL/navigation design, and email-sequence/social-content for those channels.

Input: business context, customer research, current content state, and competitive landscape - either from an existing product-marketing-context file or gathered by asking. Output: 3-5 content pillars with rationale and subtopic clusters, a prioritized list of topics (title, searchable/shareable classification, content type, target keyword and buyer stage, and the customer-research rationale behind it), and a topic-cluster map showing how content interconnects.

Points to a companion references/headless-cms.md guide covering CMS selection and editorial workflows (Sanity, Contentful, Strapi) and hands off to named sibling skills for execution across copywriting, technical SEO, AI-search optimization, programmatic content, site architecture, and email/social channels.

Content strategists and marketers deciding what to create and in what order, backed by customer research and keyword data, rather than picking topics ad hoc.