Content Keyword Analytics TF-IDF Readability API
TF-IDF term gaps, SERP content briefs, readability, near-duplicates, internal links, intent labels, lexical groups, difficulty, heading gaps, snippet widths, content decay, topic coverage, share of voice, rank changes and opportunity scores from caller-supplied texts and ranks.
Find stemmed terms common among competitor pages but missing or under-used in a draft, ranked by TF-IDF gap.
Build a content brief from caller-supplied top-ranking HTML: common headings, word-count target, questions and recommended TF-IDF terms.
Compute Flesch Reading Ease, Flesch-Kincaid, Gunning Fog, SMOG and Coleman-Liau from caller text.
Cluster near-duplicate pages with Jaccard similarity on token shingles and connected-component clustering.
Suggest internal links by unigram TF-IDF cosine similarity and extract exact or partial anchor text from target keywords.
Label keywords as navigational, transactional, commercial or informational from bundled English modifier rules and caller brands.
Group keywords that share the same Porter2 stem signature, optionally merging groups by SERP URL overlap.
Estimate keyword difficulty as 100 × mean of capped log referring-domain scores from the caller-supplied top results.
Compare a target H1–H6 outline with competitors and list common sections the target lacks.
Estimate title and meta description pixel widths with Liberation Sans advances and truncate at caller-chosen limits.
Flag pages whose clicks or impressions show a statistically significant decline with Mann-Kendall, Sen slope and windowed percent change.
Score a draft against required subtopics with synonym phrases, listing covered, partial and missing topics.
Compute estimated clicks, visibility and share of voice per domain from rank snapshots, volumes and a caller-supplied CTR curve.
Summarise winners, losers, new/lost keywords, top-3/top-10 crossings and volatility between rank snapshots.
Rank keywords by click upside to a target position, weighted by difficulty using a disclosed formula.