Topic clusters and AI: building content strategy from search data -- Copper Sun AI

Hub and spoke is an architecture decision first

A topic cluster is a structural choice before it is a creative one. The hub page earns broad authority for a subject — typically a competitive head term — by fielding the canonical overview. Spoke pages drill into every meaningful subtopic, earning long-tail traffic and then flowing link equity back up to the hub. Internal links are the mechanism; without deliberate linking, you have a collection of pages, not a cluster.

The problem with building clusters from assumptions is that you are mapping content to what you think people search, not what they actually do. Two teams building a cluster on "content marketing" will produce different spoke lists, shaped by whoever was loudest in the planning session. Both will leave real search volume untouched.

Search data as the territory map

GSC (Google Search Console) data tells you the actual terms a site already shows up for, the queries it ranks at page 2–3 (high-opportunity, not yet earning clicks), and the questions appearing in "People also ask" for your target hub term. Keyword research tools extend that picture into competitor gaps and estimated volume.

The architecture question becomes concrete: given the hub term, which spokes already exist, which are implied by search data but missing, and which are present but not internally linked properly? Brass-SEO's content architecture feature answers this by pulling the cluster picture from GSC and keyword data in one view — you see the hub, the candidate spokes ranked by opportunity, and the internal linking gaps side by side.

That inventory changes the briefing conversation entirely. Instead of asking "what should we write about content marketing?" you are asking "which of these specific search intents are we not covering, and which existing pages can we strengthen with a few new links?" Using search data to brief AI is where that inventory becomes a structured workflow for briefing each spoke.

What search intent tells you about brief shape

A topic cluster is not one type of content — it is a family of different intents held together by a shared subject. The hub page is typically informational with a long shelf life. Spoke pages land across the intent spectrum.

An informational spoke ("what is a topic cluster content strategy") needs depth, clear definitions, and internal links to every adjacent spoke. A commercial spoke ("topic cluster content strategy tool") needs a short, credible overview followed by evaluation criteria and a specific recommendation. Briefing AI the same way for both produces content that technically covers the keyword but does not convert the right intent.

The brief inputs that differ between informational and commercial spokes are not complicated, but they need to be explicit: primary intent, target funnel stage, which internal links to include, and what the reader should do next. Skipping those fields is how teams end up with a cluster that looks complete on paper but underperforms commercially. The AI brief writing guide covers how to structure those inputs so AI produces spoke content that fits its place in the cluster rather than defaulting to a generic informational treatment.

A worked example: "video content production" cluster from GSC data

Say a marketing team pulls GSC data for a site that ranks on "video content production" at position 14. They have an existing overview post — call that the hub. Search data surfaces five query groups with meaningful volume that the site has no dedicated page for:

"video content production schedule template" (informational)
"video content production cost" (informational trending commercial)
"video content production agency vs in-house" (commercial)
"how to brief a video production team" (informational, bottom-of-funnel)
"video content production brief template" (informational, high conversion intent)

Each of those is a spoke. The hub links out to all five. Each spoke links back to the hub and cross-links to adjacent spokes where the intent overlaps. The GSC data also shows the hub appearing for "video production AI tools" — not a spoke the team planned, but the data says users are arriving at this cluster with that question. That is a sixth spoke.

None of this came from brainstorming. It came from the data. The structural reasoning behind this approach covers the full hub-and-spoke architecture logic, including how to sequence spoke publication to accelerate hub authority.

Where AI fits in the execution layer

AI handles spoke content well when the brief is tight: target query, intent, word count, internal links to include, one-sentence framing for the closing call to action. It handles hub content adequately when fed existing ranking articles for competitive reference and a clear outline. What it does poorly without that structure is identify which subtopics matter — that requires search data, not language model priors about a topic.

The practical workflow: use search data to build the cluster map, surface gaps and opportunities against existing pages, then brief each spoke individually with intent-specific inputs. Copper Sun's AI content strategy process covers how to organize that pipeline so the search-to-brief handoff does not get lost between tools.

The constraint that is easiest to skip is the internal linking spec in each brief. If AI does not know which pages exist in the cluster, it cannot link to them accurately. Most content operations solve this by adding a short cluster map document to the brief as reference context — a list of existing spoke URLs with one-line descriptions. That addition alone eliminates the most common structural failure in AI-assisted topic cluster execution.

Frequently Asked Questions

How many spokes should a topic cluster have?

There is no universal number — cluster size should match search opportunity, not a template. A mature site targeting a competitive hub term might build 15–20 spokes over 12–18 months; a newer site is better served starting with 5–8 high-intent spokes and filling gaps as traffic data accumulates. Starting small with strong internal linking and clear intent differentiation beats launching 20 thin pages simultaneously.

How do you find gaps in an existing cluster?

Pull GSC data for your hub page and look at the queries it shows impressions for that have no dedicated spoke — those are gaps the cluster is already being surfaced for but not properly serving. Keyword tools extend the picture to competitor rankings on subtopics you have not touched. A third category, pages that exist but are not linked into the cluster properly, surfaces through a crawl-based audit — Brass-SEO's internal linking analysis covers the mechanics of that audit and what to look for beyond raw link count.

What does AI do well for hub posts versus spoke posts?

For hub posts, AI is strongest at synthesizing competitive content into a structured outline and drafting the definitional sections — the parts that require coverage breadth rather than original research. For spoke posts targeting specific queries, AI works best with a tight brief: exact query, intent type, word count, cross-links to include, and a clear done condition. The informational-versus-commercial intent distinction is one humans need to supply explicitly; without it, AI defaults to informational framing regardless of what the data says the page should accomplish.

Does topic cluster strategy still apply when AI overviews are absorbing search clicks?

AI overviews shift the value of top-of-funnel definitional content — zero-click answers reduce traffic to pages that answer broad "what is" queries. The cluster structure still matters because commercial-intent spokes, comparison pages, and high-specificity long-tail posts are less affected by overview displacement. Building clusters with deliberate intent diversification rather than stacking informational spokes makes the overall content investment more resilient to how search result pages continue to evolve.