Authority Blueprint Builder
Enter a main keyword to generate a complete content authority structure—pillar page, topic clusters, supporting articles, internal linking patterns, and publish-order strategy.
Enter a broad topic or keyword to build your full content authority map.
What the Authority Blueprint Builder Does
The Authority Blueprint Builder analyses a target keyword and generates a complete content architecture: a pillar page, topic clusters, and supporting articles — each assigned a content type, publish priority, and internal linking plan. The result is a structured, publish-ready roadmap that reduces guesswork and prevents content overlap before a single article is written.
The tool applies intent detection to every title it produces, matching page types to actual search behaviour rather than defaulting to a single format. Informational, commercial, troubleshooting, and transactional signals are each handled differently, so the resulting blueprint reflects the real distribution of queries in a topic space.
Content diversification is handled through a deterministic rotation engine. The tool tracks which content types have been published recently — guides, checklists, case studies, troubleshooting pieces, and expert perspectives — and enforces minimum ratios across a rolling window. This prevents topical monotony and signals broad coverage to search engines evaluating site authority.
Internal linking decisions are derived from semantic overlap between the new article and the existing content inventory. The tool selects 3–7 relevant targets per new piece and distributes anchor types (partial match, contextual, branded, implicit) to avoid over-optimised patterns that attract scrutiny.
An entity coverage layer cross-references required terms against the generated outline before finalisation. If essential entities are missing from the structure, the outline is revised to include them, improving semantic completeness and reducing the risk of thin coverage signals.
Cannibalization detection runs before every blueprint is confirmed. The tool compares the candidate title, keyword, and intent against every previously generated blueprint using a weighted similarity model. Topics with high overlap receive a risk flag or are automatically pivoted to a differentiated angle, protecting the existing content investment.
Headline generation follows documented patterns calibrated to each content type, with a rule that prevents the same structural template from appearing in consecutive posts. Robotic formulas — "How to X Step by Step", "Top 10 Tips for Y" — are blocked entirely. Every pillar title is drawn from a pool of contextual candidates rather than a fixed template.
Frequently Asked Questions
How does the tool prevent keyword cannibalization?
Before each blueprint is finalised, the tool compares the candidate pillar title, its keyword, and its search intent against every previously generated entry in the publish history. The comparison uses a weighted Jaccard similarity model across three dimensions: title overlap (50%), keyword overlap (35%), and intent match (15%). Topics that exceed the risk threshold receive a recommendation to pivot angle, merge with the existing piece, or update the live version rather than publish a duplicate.
How are case studies rotated into the content mix?
The rotation engine tracks the last 20 published pieces (configurable) and enforces a minimum 15% case study ratio within that window. If three consecutive non-case-study pieces are detected, the next content type is forced to case study regardless of the keyword signal. The rules are deterministic — there is no randomness — so the rotation pattern is predictable and auditable.
Does the tool analyse search intent?
Yes. Every title produced by the blueprint — pillar, cluster, and article — is assessed against keyword signal lists for transactional, commercial investigation, informational, and navigational intent. The detected intent type influences the suggested word count range, publish priority score, and which snippet blocks (definition, steps, FAQ) are recommended for the outline.
How does the tool select internal link targets?
Internal links are selected by computing semantic similarity between the new article's keyword and every existing cluster and article title in the current blueprint. The top 3–7 most relevant pieces are selected, and each is assigned a different anchor type — partial match, contextual, branded, or implicit — to maintain a natural distribution. Reverse-link suggestions (existing articles that should link back to the new piece) are generated at the same time.
What makes the headlines less robotic?
Each content type — guide, checklist, case study, troubleshooting, expert opinion — has its own headline template pool. Structural patterns like "How to X Step by Step" and "Top 10 Tips for Y" are explicitly banned. A pattern-breaker rule prevents the same structural template from appearing more than once in the last five generated titles. The selected pillar title is always drawn from the human headline pool, never from a fixed default string.
Is historical data stored between sessions?
Yes. The publish history is persisted to localStorage under the key authority_history. On every page load, the history is restored from storage, so rotation rules, cannibalization checks, and freshness flags all reflect the full accumulated record rather than starting from zero. If localStorage is unavailable (private browsing, storage quota exceeded), the tool falls back silently to an in-memory session log.
Can I adjust the content diversification ratios?
Yes, through the internal settings object. The configurable parameters include rotation_window (how many recent pieces to evaluate), case_study_ratio_target (minimum case study share), and force_case_study_after_n_non_cs (consecutive non-case-study threshold). These can be updated at runtime via BlueprintLogic._internal.updateSettings({...}) without reloading the page.
Does the tool modify my site design or inject anything into my pages?
No. The tool generates a structured data blueprint — titles, sections, intent types, word count ranges, linking plans — that informs your editorial process. It does not connect to any CMS, modify any live pages, or insert content automatically. All output is advisory and export-only (JSON, CSV, or plain text report).