Keyword Intake & Intent Mapper (Raw List → Intent Buckets → Topic Map)
Quickly turn a messy keyword list into clear intent buckets and starter clusters you can sort, score, and brief.
Prompt Overview
Tips For You
- If your list is huge, run this in batches by theme or product line. - Add available metrics as extra columns; the model will preserve them. - Include business positioning so titles sound like you, not generic SEO. - Keep cluster sizes tight to simplify briefs and internal links.
From Operations TeamNexusAi TechnologyProblem It Solves
Reduces manual sorting, inconsistent naming, and unclear search intent that slow down planning.
Intent detection
Classifies each query by primary search intent to guide content type.
Topical clustering
Groups semantically related keywords into manageable clusters.
De-duplication
Merges near-duplicates to reduce wasted effort and cannibalization.
Working titles
Generates draft titles aligned to audience language and SERP cues.
AI Prompt Instructions
Act as: A senior SEO content strategist specialized in search intent mapping, topical clustering, and editorial planning.
Why this task matters: Raw keyword lists are noisy. Mapping search intent and organizing into clusters creates the foundation for briefs, internal links, and a realistic calendar.
Important boundaries:
- Avoid hallucinating volumes or difficulty—use only provided metrics or placeholders.
- No duplicate topics; de-duplicate close variants.
- Keep cluster names human-readable and strategy-friendly.
User inputs (paste below this prompt):
- Business context: industry, audience, products, geos.
- Raw keywords (CSV or lines) with any available metrics (volume, CPC, difficulty, URL, notes).
- Publishing cadence target (e.g., 2 posts/week).
Objectives:
1) Classify each keyword by primary intent (Informational, Commercial, Transactional, Navigational) and secondary modifiers.
2) Group into topical clusters with a pillar → subtopic structure.
3) Suggest canonical target keyword per cluster and early working titles.
4) Flag obvious duplicates and merge.
Analysis workflow:
1) Normalize inputs (lowercase, trim, remove punctuation that doesn’t change meaning).
2) Detect intent using query verbs, modifiers (best, vs, how, price), and SERP patterns (generic, brand, location, product).
3) Build clusters by semantic proximity and overlapping modifiers; cap clusters at 8–12 items for manageability.
4) Propose a pillar topic per cluster and map subtopics beneath it.
5) Create working titles optimized for clarity and click-worthiness, not keyword stuffing.
6) Produce a topic map with: cluster_id, cluster_name, target_kw, intent, subtopic_kw, suggested_title, duplicate_of (if any), notes.
Required output format:
- Provide a concise summary first (3–5 bullets), then a CSV table with headers: cluster_id,cluster_name,target_kw,primary_intent,subtopic_kw,suggested_title,duplicate_of,notes
Quality controls:
- Each subtopic must roll up to a clear pillar.
- No cluster name should be a single keyword; use a descriptive phrase.
- Titles must be unique across the table.
Verification checklist:
- Duplicates merged?
- Ambiguous intent resolved with rationale?
- Pillar topics broad enough to support internal links?
Final instruction: Print the summary and full CSV table only. Do not include extra commentary beyond the requested sections.
Expected Outcome
Summary: - 7 clusters formed across informational and commercial intents - 64 subtopics de-duplicated; 9 merged - Clear pillar → subtopic structure ready for briefs CSV: cluster_id,cluster_name,target_kw,primary_intent,subtopic_kw,suggested_title,duplicate_of,notes C1,Content Calendar Basics,content calendar template,Informational,seo content calendar,"SEO Content Calendar Template: Build a 90-Day Plan",,High search volume theme C1,Content Calendar Basics,content calendar template,Informational,editorial calendar,"Editorial Calendar vs Content Calendar: What’s Different?",,Clarify terminology
Implementation Journey
Map intents and clusters in ChatGPT
Open ChatGPT. Paste business context and your raw keyword list. Ask for intent classification and topical clusters using the prompt. Expect a summary and a CSV-like table. Copy the table output.
15-25 minStructure the table in Google Sheets
Paste the CSV table into Google Sheets. Create filters for primary_intent and cluster_id. Add helper columns for owner and due_date. This prepares the data for scheduling and handoff.
10-15 minPublish the topic map to your planning tool
Import the sheet to Notion or Asana. Create a board grouped by cluster_id. Assign owners and set due dates to match your cadence. This becomes your working topic backlog.
10-20 min
