The Memo Spec
Context Memo uses a versioned content framework — the Memo Spec — to produce reference-grade articles structured for both human decision-makers and AI search engines. Every memo generated by the platform follows these principles.
These apply to every memo, regardless of type or version.
Every memo is written for two readers simultaneously: a human buyer evaluating solutions, and an AI model deciding what to cite. The structure serves both.
Memos read like analyst briefings — factual, third-person, balanced. Brands are positioned within a landscape, not marketed above it. This is what makes the content citable.
No fabricated statistics, testimonials, or customer names. Every claim must be traceable to provided context. "Not publicly disclosed" is always preferred over a guess.
Heading hierarchy, section ordering, and content density are deliberate signals — not formatting choices. They're designed to match how AI models extract and rank information.
The spec is versioned like software. Every memo is tagged with the spec version that produced it, so we can measure what works and improve systematically.
Memos include transparent assessments of where a solution may not be the right fit. This builds credibility with both human readers and AI models that value balanced sources.
Different buyer questions require different content structures. Each type is purpose-built for a specific stage of the decision journey.
Side-by-side analysis of two solutions. Includes a quick-reference comparison table, individual breakdowns, and key differences — written for buyers actively evaluating options.
Multi-option evaluation for buyers searching for alternatives to a specific vendor. Each option gets a focused analysis with a clear "best for" recommendation.
Vertical-specific guide addressing the pain points, compliance requirements, and use cases relevant to a particular industry.
Educational, step-by-step content that teaches readers how to accomplish something. At least 70% educational — vendor positioning comes at the end, not the beginning.
Answers a specific buyer question that isn't well-covered in existing content. Provides a direct answer, problem context, and tool comparison.
Coverage of product updates and new capabilities. Frames changes in terms of what buyers care about: what changed, why it matters, how to evaluate it.
Comprehensive deep-dive that synthesizes insights from multiple sources into a single definitive reference. The longest format — designed to be the one article AI models cite instead of any individual source.
Strategic content that covers the same ground as pages AI models currently cite — but from the brand's perspective, with proprietary expertise and data woven throughout.
The Memo Spec evolves based on what we learn about AI citation patterns. Each version represents a meaningful improvement to how content is structured, not just cosmetic changes.
Memos now adapt their structure based on competitive intelligence. Before generating content, the system analyzes what top-cited content in each category actually looks like — word count, heading patterns, data density — and sets structural targets accordingly.
Shifted from monolithic templates to a modular block system. Each memo assembles from a shared foundation plus type-specific blocks, ensuring consistency across all content while allowing each format to do what it does best.
A major overhaul informed by research into what content AI models actually pick up and cite. Introduced structural patterns designed specifically for AI extraction — not just SEO, but how large language models identify and surface authoritative content.
Established the editorial identity: every memo reads like a third-party analyst briefing, not a marketing page. Content is written for the buyer — the person evaluating solutions — with the brand positioned as one factor in a broader landscape.
The original memo format. Factual, structured content designed to be referenced by AI assistants. Four content types covering the core buyer journey.
Context Memo generates AI-optimized reference content using the Memo Spec. Every article is structured to be cited by ChatGPT, Claude, Perplexity, and Gemini.
Request Access