Blog Cluster Plan Template
Stop publishing one-off posts. Cluster your content around a pillar and the whole cluster compounds for SEO and for AI citations.
Why this matters
- 1Topic clusters beat one-off posts because Google ranks the pillar, then ranks every spoke in its slipstream. One pillar can lift 10+ posts at once.
- 2AI engines (ChatGPT, Claude, Perplexity) cite the most thorough source on a topic. A pillar plus spokes is thorough by design.
- 3Internal linking is the single biggest on-page ranking factor that you fully control. Clusters force good internal linking.
- 4A cluster gives you 8 to 12 weeks of editorial calendar from a single planning session.
The template, step by step
- 01
Pick a pillar topic broad enough to support 10+ spokes
Good pillar topics are noun-shaped umbrellas: 'agentic AI', 'AI SEO', 'AI CRMs', 'topic clusters'. Bad pillar topics are too narrow ('Claude Opus 4.7 prompts') or too broad ('marketing'). The test: can you list 10 distinct questions a reader might have about this pillar? If yes, it's a pillar.
- 02
Brainstorm 12 to 15 spoke questions
Cast wide on the first pass. Mine People Also Ask, Reddit, Twitter searches, and your own customer support tickets. Each spoke should be a specific question (e.g. 'how to write llms.txt') not a topic (e.g. 'llms.txt'). Questions are easier to rank for because they match search intent precisely.
- 03
Cluster spokes into 3 to 4 sub-themes
Sort the 12-15 spokes into groupings. 'What it is' spokes, 'how to do it' spokes, 'comparison' spokes, 'tools/templates' spokes. The sub-themes become H2 sections in your pillar page and become the navigation logic on the pillar.
- 04
Write the pillar page last, not first
Counterintuitive but right. Write 4-6 spokes first. As you write them, you discover the natural shape the pillar should take, the questions to answer, the language to use. The pillar becomes a synthesis instead of an outline you have to fill in.
- 05
Internal-link every spoke to the pillar and to 2-3 sibling spokes
Every spoke gets at least one link to the pillar (helps the pillar rank). Every spoke gets 2-3 links to sibling spokes (helps spokes rank). Use descriptive anchor text, not 'click here.' AI engines weight anchor text heavily for understanding topic relationships.
- 06
Add FAQPage schema to every spoke
4-6 FAQ Q&As at the bottom of each spoke. These get pulled into Google's AI Overview and Perplexity citations more than any other content type. Use real questions you've answered in customer support, not invented ones.
- 07
Ship one spoke per week, pillar at the end
8 to 12 weeks of weekly cadence. Don't drop them all at once: Google indexes faster with regular cadence than with bulk dumps. The pillar ships last and ties everything together. Update spokes monthly with new info to signal freshness.
PILLAR TOPIC: ____________________________
SUB-THEMES (3-4):
Theme A: ____________
Theme B: ____________
Theme C: ____________
Theme D: ____________
SPOKE QUESTIONS (12-15):
Theme A spokes:
1. ____________
2. ____________
3. ____________
Theme B spokes:
4. ____________
5. ____________
6. ____________
Theme C spokes:
7. ____________
8. ____________
9. ____________
Theme D spokes:
10. ____________
11. ____________
12. ____________
PILLAR OUTLINE (write after 4-6 spokes are live):
H1: [pillar topic, definition framing]
H2 per sub-theme, with links to all spokes in that theme
FAQ section at bottom
CTA above fold + bottom
EDITORIAL CADENCE:
Week 1: Spoke 1
Week 2: Spoke 2
...
Week 10: Spoke 10
Week 11: Pillar page goes live
Week 12: Update first 2 spokes with new infoCommon mistakes
- Picking a pillar that's too narrow. If you can't get to 10 spokes, the pillar isn't a pillar, it's a spoke.
- Writing the pillar first. You don't know what to put in it until you've written some spokes.
- Skipping the internal linking pass. A cluster without internal links is just 12 unrelated posts.
- Dropping all 12 spokes at once. Google rewards consistent cadence over bulk uploads, especially for new domains.
- Treating spokes as throwaway. Each spoke should rank on its own merit. The cluster is the bonus, not the only point.
Let PostWyse run this template on autopilot.
Wyse drafts every input, every personalization, every follow-up in your brand voice. You approve before anything goes live.
Questions
How many spokes does one pillar need?
Minimum 8, ideal 10-12, max 15. Below 8 and the pillar doesn't have enough internal-link weight to lift. Above 15 and you're padding with weak spokes that drag the cluster's average quality down.
Can a spoke belong to two clusters?
Yes, but only one is its primary cluster. The primary cluster's pillar gets the strongest link from that spoke. The secondary cluster gets a regular contextual link. Trying to make a spoke fully serve two pillars usually means it serves neither.
How often should I update the pillar after publishing?
Quarterly at minimum. Add a 'last updated' line at the top of the pillar. Refreshing the pillar signals freshness to Google and re-crawls the whole cluster. AI engines also weight recency signals heavily for citation selection.
What's the right word count for a pillar page?
2,500 to 4,000 words. Below 2,500 it can't synthesize 10+ spokes meaningfully. Above 4,000 and readers stop scrolling, dropping engagement signals. The pillar should feel like a definitive guide, not a textbook.
Does this work for AI engines as well as Google?
Yes, arguably better. AI engines (ChatGPT, Claude, Perplexity) are explicitly looking for the most authoritative source on a topic, which is exactly what a well-built cluster signals. The pillar gets cited as the canonical answer, the spokes get cited for specific sub-questions.