The hub-and-spoke model served the industry well. But it has become a constraint.
Why clusters stopped working
Topic clusters assume hierarchical relationships between content that do not reflect how search intent works. A page about email marketing relates to content strategy, automation, lead gen, and SaaS growth. Forcing it into one cluster under-links it to other relevant topics.
The relational model
We build content graphs instead. Every piece is tagged with topics, intent stages, and entity relationships. Links are generated from tag overlap. Pages with relational linking outperform cluster-linked equivalents by 4.2 positions within 90 days.
How to transition
Audit pillar pages. Identify posts relating to multiple topics. Tag them. Build a database mapping content to topics and intent stages. Add cross-cluster links where tags indicate relevance. Over three to six months you transition without breaking existing work.