Cluster keywords into a page map
Turn a keyword set into defensible page decisions. Do not group terms only because they share words.
Invoke Rankizz MCP tools
When using Rankizz through MCP, the initial tool list is intentionally compact. Call search_tools with a capability query or exact domain-tool name and set includeSchemas: true unless the arguments are already known. Invoke a discovered tool with call_read_tool when its readOnlyHint is true; use call_tool for every other tool. Pass the discovered name as tool and its original input object as arguments. REST callers continue to use the direct /api/v1/tools/<tool-name> endpoint.
Prepare the input
Accept CSV, spreadsheet, Markdown, JSON, copied text, saved Rankizz keywords, or a prior research result. Preserve:
- original keyword
- volume, difficulty, CPC, rank, or intent when supplied
- source and retrieval date
- existing tags or proposed page
Normalize casing and whitespace for comparison without destroying the original values. Separate branded, irrelevant, wrong-market, and malformed terms before clustering.
Build evidence locally
Use the site's existing URL inventory, navigation, sitemap, or supplied page list to understand current page purposes. Rankizz is not required for semantic and intent clustering when the input already contains enough context.
For ambiguous or high-value groups, use live SERP similarity if available. Similar wording does not prove the same intent; different wording does not prove separate pages.
Use Rankizz conditionally
When connected:
- call
list_saved_keywordsto load a reviewed project set - call
get_keyword_metricsonce for important terms missing metrics - call
get_serp_resultsin batches for ambiguous clusters - call
get_site_explorer_inventoryto map clusters to known site URLs - call
get_ranked_keywordsor Search Console tools when existing performance affects the mapping
Rankizz-hosted calls require at least 10 credits. On a low-credit response, stop cloud calls, link to https://www.rankizz.com/billing, and continue with available inputs.
Do not save keywords, replace tags, or reorganize project data without explicit confirmation.
Decide page boundaries
Follow references/clustering-rules.md.
For each cluster, decide whether to:
- map to one existing page
- improve or consolidate existing pages
- create one new page
- split into multiple page intents
- retain for validation
- exclude
Prefer the searcher's expected page and task over a mechanically convenient taxonomy.
Deliver
Return:
- a clustering summary and evidence limitations
- the proposed page map
- cannibalization and consolidation findings
- excluded or uncertain terms with reasons
- a prioritized creation or refresh queue
Each cluster must include primary keyword, secondary terms, intent, recommended page type, target URL or proposed slug, aggregate demand only when mathematically valid, evidence, and confidence.
Do not sum duplicate or close-variant volume as if every value were independent. Do not claim SERP similarity unless SERPs were actually compared.
- Downloads
- 75All time
- Security audit
- Pass
- Last updated
- 2d ago
- Current version
- v1.0.0
- License
- MIT-0
