Runs an existing page through a four-stage query fan-out analysis to find every related question it should be answering but isn't.
- Inputs: existing content, primary topic/keyword, target audience, content goal
- Extracts entities and themes, then maps likely user queries per theme
- Scores each query as fully, partially, or not addressed, with a 1-10 depth rating
- Splits gaps into critical, opportunity, and enhancement tiers
- Rebuilds the piece as a 20-25 heading outline flagging new gap-filling sections
Diagnoses why a specific CTA or form is underperforming by reviewing the current CTA copy, form fields, funnel steps, and any drop-off data you can supply.
It returns three rewritten CTA variations per button, a keep/remove/modify audit of each form field, micro-copy fixes, a friction score per element, progressive disclosure ideas, and an expected conversion lift estimate. Works even without drop-off data, since it still flags obvious friction from the funnel description alone.
Scores a piece of content against Google's Experience, Expertise, Authoritativeness, and Trustworthiness framework, using the content itself plus author bio and website context as inputs. It returns a scorecard rating each E-E-A-T element as present, weak, or strong with specific evidence pulled from the text, a ranked list of priority improvements, and concrete implementation examples for the top three fixes, such as adding a credentialed author bio. Apply the stricter YMYL standard whenever the topic touches health, finance, or legal advice.
A pre-publish QA pass for AI-assisted or human drafts, checking the things a spellchecker won't catch.
- Inputs: draft content, target keywords, brand guidelines, and source materials to fact-check against
- Flags potential AI hallucinations, unsupported claims, and factual inaccuracies
- Runs a citation audit for missing or needed sources
- Checks style consistency and brand voice alignment
- Outputs an originality score with flagged passages plus a prioritized list of revisions
- Run in deep-research mode so claims get checked against the actual source materials
Compares your draft against every article currently ranking on page one of Google for a target keyword, rather than guessing at what's winning.
It surfaces the ranking factors shared across the top results, the specific content gaps and depth differences between your piece and theirs, and any unique elements like tools, media, or data you're missing, then delivers a priority-ordered improvement roadmap. Needs a model that can actually browse the live SERP.
Handles heading structure at both ends of the writing process: give it a guide topic and the key points it needs to cover before you write, and it proposes an SEO-friendly H1/H2/H3 structure to draft against. Paste in a finished draft afterward along with the primary keyword, and it rewrites the existing headers to better target that keyword while keeping the structure scannable. Useful for planning long-form guides and for cleaning up headers on drafts written without much structure in mind.
Generates a spread of title tag options for one topic, each built around a different psychological trigger so you can test angles instead of guessing at one headline.
- Inputs: topic/keyword and target audience
- Produces titles across ten frameworks: benefit-driven, curiosity gap, authority, urgency, problem-agitation, contrarian, social proof, how-to, list, comparison
- Keeps every title under 60 characters with the keyword worked in naturally
- Output: a table with framework, title, character count, emotional trigger, best-use case
Suggests internal links for a specific article using a site URL inventory (URLs plus their headlines), the article's own URL, and its full text, focused on sentence-level relevance rather than keyword matching.
Each recommendation includes the exact source sentence, a natural anchor text that still makes sense read out of context, the target URL, and the reader benefit of clicking, plus a validation checklist per link. Reject any suggestion that fails the 'makes sense out of context' test.
Roots a keyword or content strategy in Clayton Christensen's Jobs-To-Be-Done framework instead of raw search volume, starting from a single seed topic.
- Input: just a seed topic
- Identifies the functional, emotional, and social jobs customers are trying to accomplish
- Translates each job into 3-5 real search queries, problem-focused versus solution-focused
- Maps every query to a funnel stage: awareness, consideration, decision, retention
- Output: a JTBD query map table plus a job prioritization matrix and content gap analysis