Topic Cluster Builder & Keyword Expansion System
A systematic approach that breaks down a complex process into actionable steps for consistent results.
Use This When
Articles, service pages, AEO/GEO content, interlinking, SERP-informed briefs.
Inputs Needed
URL, target keyword, audience, competitors, location, search intent, internal links, products/services.
Expected Output
SEO brief or content draft with search intent, outline, on-page elements, internal links, FAQ, schema suggestions.
The Workflow Prompt
You are a technical SEO strategist and editorial content lead. Objective: Topic Cluster Builder & Keyword Expansion System Context: A systematic approach that breaks down a complex process into actionable steps for consistent results. Original task: You are a content strategist who has engineered topic cluster systems for 300+ content properties, improving organic traffic by 3-10x through comprehensive keyword and topic coverage.Create a comprehensive topic cluster building system for [MAIN TOPIC]. This system must include:1. Core topic definition clearly articulating the main topic and primary audience search intent2. Keyword research and expansion methodology using tools and search analysis to identify 50-200 related keywords and questions3. Search intent classification organizing keywords by intent type (informational, navigational, transactional, commercial)4. Topic clustering algorithm grouping related keywords into logical content clusters5. Cluster naming and organization creating logical groupings with clear topic labels6. Content gap identification - identifying which clusters have content coverage and which represent content opportunities7. Content prioritization determining which clusters to address first based on search volume, relevance, and competition8. Internal linking structure mapping how content pieces within a cluster connect and how clusters link to main pillar9. Keyword-to-content assignment ensuring each content piece targets specific keyword clusters10. Topic cluster evolution - adding new clusters as audience needs evolve and new questions emergeInclude cluster maps, keyword organization examples, and traffic impact examples. Inputs I may provide: URL, target keyword, audience, competitors, location, search intent, internal links, products/services. Operating instructions: - First, restate the objective in one clear sentence. - If critical information is missing, ask up to 5 focused questions. If there is enough information to proceed, make practical assumptions and label them. - Use a Detailed response style. - Be specific to the business, audience, channel, and constraints provided. - Avoid generic AI advice. Give concrete recommendations, examples, templates, copy, or steps I can use. - When current facts, competitors, laws, prices, policies, or market claims matter, use current research and cite sources. - Do not expose hidden chain-of-thought. Provide a concise rationale or decision summary instead. - End with a short QA checklist that helps me verify the output. Required output: SEO brief or content draft with search intent, outline, on-page elements, internal links, FAQ, schema suggestions. Caution: Avoid generic output; require concrete examples, assumptions, and next steps.
QA Follow-Up Checklist
After the AI returns its output, verify against:
- Output is specific to the provided business/context.
- Assumptions are clearly labeled.
- No unsupported claims without source checks.
- Next actions are clear and usable.
- Search intent, internal links, FAQ, and on-page elements are included.
Follow-Up Prompt
Now turn the result for 'Topic Cluster Builder & Keyword Expansion System' into a client-ready version: tighten wording, remove fluff, add missing assumptions, and provide the next 3 actions.
Avoid / Cautions
Avoid generic output; require concrete examples, assumptions, and next steps.
How Different Verticals Use This Workflow
Restaurant & Hospitality
A boutique hotel group with 4 properties in Charleston feeds in their booking page URLs, their main competitor (a larger Marriott-affiliated chain), and the seed topic 'historic Charleston accommodations'. Output: 3 pillar clusters — historic district neighborhoods, weekend itineraries, and event venues — with 18 spoke articles mapped to specific booking pages. Outcome: 47% organic traffic increase to booking pages in 6 months. The plan kills generic 'best hotels in Charleston' content because three review sites own it.
Retail & E-commerce
A Shopify store selling Japanese kitchen knives, $800K ARR, feeds in their PDPs, their top competitor (Knifewear), and a seed topic of 'gyuto knives'. The cluster maps 4 pillars: knife types, steel composition, care, and chef stories. 24 spoke articles each link to specific product collections. Outcome: 3 spokes hit page 1 within 4 months, $11K incremental monthly revenue tied to organic. Skips the saturated 'best chef knife' query in favor of 'gyuto vs santoku for home use'.
Professional Services & B2B
A SaaS company selling payroll software for restaurants, $3M ARR, feeds in their feature pages, their competitor Toast Payroll, and seed topic 'restaurant payroll'. The cluster builds 3 pillars: tip reporting, multi-state compliance, and turnover. 19 spokes each tie to a free tool or template lead magnet. Outcome: 142 demo requests from organic in quarter 1 vs 38 the prior quarter. The plan rejects 'what is payroll' style top-of-funnel content as too far from purchase intent.
Beauty & Personal Care
A clean skincare DTC brand selling vitamin C serums, $2.4M ARR, feeds in their 6 product pages, competitor Drunk Elephant, and the seed 'vitamin C serum'. The cluster builds 2 pillars: ingredient science and skin concerns. 14 spokes each link to specific products tied to skin concern. Outcome: 31% organic traffic to PDPs in 90 days. Plan refuses to write 'best vitamin C serum 2026' because affiliate sites own it and trying to compete burns budget.
Local & Trade Services
A roofing company doing $4M/year in metro Phoenix feeds in their 8 service pages, their competitor RoofMaxx, and the seed 'tile roof repair Phoenix'. The cluster maps 3 hyper-local pillars: tile roofs, monsoon damage, and HOA-compliant repairs. 16 spokes each include a city neighborhood modifier. Outcome: 23 inbound leads/month from organic, up from 6. The plan ignores national keywords because local intent is the whole revenue engine.
Frequently Asked
What inputs actually move the needle for a topic cluster build?
Your existing top-10 organic pages with their current keyword rankings, the three competitors who outrank you, and a non-negotiable 'what we sell' statement. Without the sales tie-in, the cluster expands into informational keywords that bring traffic but zero revenue. Also feed in your Search Console data for the last 90 days — the queries you almost rank for are gold and most cluster prompts miss them. Don't feed in 'industry trends'. That's how you end up with a cluster about 'the future of [thing]' nobody searches for.
How is this different from just running keyword research in Ahrefs?
Ahrefs gives you a list. This prompt gives you a content architecture: pillar page plus spokes, internal link map, and which clusters to build first based on commercial intent and your existing topical authority. Keyword research is the input — clustering is what determines whether you build one strong pillar or thirty thin pages that cannibalize each other. Most teams build the thirty pages because the keyword tool didn't tell them to consolidate. Use the prompt after the keyword export, not instead of it.
What's the most common failure mode when running this?
Over-clustering. The model wants to give you 12 pillar topics because you fed it 200 keywords. You can't actually publish 12 pillars in a quarter, so you ship none of them. Force the prompt to recommend only 3 pillars max for the first 90 days and rank them by 'how close you already are to ranking'. The second failure: ignoring intent — putting transactional and informational keywords in the same cluster, then writing one page that ranks for neither.
Should I run GPT-5.5 or pair it with Perplexity?
Pair them. GPT-5.5 alone hallucinates search volumes and competitor URLs. Use Perplexity Sonar to confirm which competitors actually rank for the cluster terms today, then hand that real data back to GPT-5.5 to build the architecture. Skip this pairing only if you're working in a niche where you already know the top 10 ranking URLs cold. For 90% of clients, the Perplexity pass catches three competitors GPT didn't know existed.