Build the Cluster, Not the Post
A 12-week field record from Google Search Console: how building one interlinked content cluster took a topic from about 14 to 1,174 weekly clicks.
For about two years, one Gourmet Host article on after-dinner speeches did something quietly useful. It held a small, steady trickle of organic clicks while nothing was done to it. Low, but never zero. That trickle was the tell: people were searching for help with speeches and toasts, and a single ageing page was catching the overflow.
The question that started this note was simple. What happens if you stop treating that as one page and start treating it as a topic?
Here is the trajectory, measured in a rolling 28-day window sampled each week.
Key takeaways
- One quiet, low-but-steady page is a signal of unmet demand, not noise.
- Building the topic as a deep, interlinked cluster took it from about 14 to 1,174 organic clicks per 28-day window, roughly 80x in 13 weeks.
- The last page added ranked fastest, because the cluster around it had already earned the authority it inherited.
The starting point
Through most of the window the topic was flat. Across the back half of 2025 and into February 2026, the speeches and toasts pages on the site pulled between roughly 4 and 15 organic clicks per 28-day window, combined. The 28 days ending March 15, 2026 sat at 14. That is the baseline: a real, persistent, small signal that had been there long enough to trust.
A flat line for ten months is where most sites give up on a topic. The steady trickle says the opposite. Demand that survives neglect is demand worth building for.
What we did: two deliberate waves
This was not one big publish. It was two waves, built on different logic.
| Wave | When | Pages | The thinking |
|---|---|---|---|
| Wave 1 | mid-March 2026 | ~14 | Build out the human element around speeches, toasts, and event communication. Timed deliberately, well before wedding season peaked, so the pages would be ranked and ready when demand arrived. |
| Wave 2 | mid-May 2026 | ~10 | Triggered by Wave 1’s early success and a second insight: a blog audience is a reading audience, and a committed group wants this content in written form. Wave 2 was keyword-led, researching a wider set of high-volume, low-competition terms the site could realistically win. |
The cluster went from one quietly-surviving page to roughly 24 interlinked pages in about eight weeks, each one pointing to the others so the whole topic read as a single, deep resource rather than scattered posts.
What the numbers did
The topic moved the way a cluster moves: slow to start, then sharply, then compounding. Each figure below is the trailing 28-day click total as of that Sunday.
| 28 days ending | Organic clicks (28-day) | Note |
|---|---|---|
| Mar 15, 2026 | 14 | baseline, pre-build |
| Mar 22, 2026 | 46 | Wave 1 pages enter the data |
| Mar 29, 2026 | 113 | |
| Apr 12, 2026 | 252 | |
| May 10, 2026 | 362 | Wave 2 pages enter the data |
| May 24, 2026 | 694 | |
| Jun 07, 2026 | 1,077 | |
| Jun 14, 2026 | 1,174 | 24 pages, avg position 8.6 |
That is roughly 80x the pre-build window, across thirteen weeks. And people read what they found: in the 28 days ending June 14, the cluster drew 1,261 organic sessions and 1,454 page views in GA4, which is the point of writing for a reading audience.
The tell within the data: the cluster lifted the pillar
The clearest proof that this is a cluster effect, and not just more posts, is the page that arrived last.
The father-of-the-groom-speech page was a Wave 2 page. It first showed up in the data in the window ending May 10, and its 28-day average position was already on page one a week later. It now pulls 409 clicks per 28 days on its own at position 6.5, the single biggest page in the cluster. A brand-new page does not normally rank that fast. It ranked fast because the cluster around it had already taught Google that this site is a credible source on speeches and toasts. The new page inherited that authority.
| Page | Wave | First seen | Reached page 1 | Clicks (28d) | Pos |
|---|---|---|---|---|---|
| father-of-the-groom-speech | 2 | May 10 | May 17 | 409 | 6.5 |
| anniversary-speeches-and-toasts | 1 | Mar 22 | Mar 29 | 188 | 7.8 |
| anniversary-toasts | 1 | Mar 22 | Apr 19 | 134 | 5.9 |
| funny-jokes-for-toasts | 1 | Mar 22 | Mar 22 | 131 | 8.1 |
| best-farewell-speech-ideas | 1 | Mar 22 | Mar 22 | 127 | 8.3 |
| best-quotes-for-toasts | 1 | Mar 22 | Mar 22 | 104 | 8.3 |
| father-to-daughter-wedding-speech | 1 | Mar 22 | climbing | 41 | 10.3 |
And the spread of terms it now earns shows the keyword-led second wave paying off. The cluster ranks not for one phrase but for a long family: father of the groom speech and a dozen close variants, plus anniversary toasts, funny toast, toast jokes, anniversary speech, and farewell speech. Breadth of winnable terms is what a cluster buys you.
What this teaches
- A quiet page is a signal, not noise. A page that holds low-but-steady traffic with no effort is unmet demand showing itself. Read your oldest, dead-but-not-dead pages as a map of where to dig.
- Depth is the unit of SEO, not the article. One page sat flat for ten months. About 24 interlinked pages moved the topic 80x. You do not rank a post; you rank a topic, and topics are won with depth.
- Interlinking makes the whole lift together. The Wave 2 pillar reached page one in a week because Wave 1 had already built the topical authority it inherited.
- Build ahead of demand. The cluster was built before peak season, so it was already ranked when seasonal search arrived. Ranking has lag; plan for it.
- Let early wins fund a sharper second wave. Wave 2 was bigger and more keyword-led precisely because Wave 1 proved the topic. Validate, then expand with research.
- Match the medium to the audience. A blog audience is a reading audience. The bet was that a committed group wants this content in written form. The 1,261 sessions per 28-day window are that bet, confirmed.
What this does not claim
Honesty keeps the note useful. Spring and early summer lift demand for speeches and toasts on their own, so seasonality is part of this curve, not separate from it. The whole site’s authority was also growing across this window, which helps every page, not only these. And organic clicks are a visibility measure, not a revenue figure. The cluster build is the lever we are pointing at. It is not the only force in the room. What the data does support is the shape: a topic that sat flat for ten months moved sharply once it was built with depth and links, and the last page in lifted fastest because the cluster carried it.
Frequently Asked Questions
What is a content cluster?
A content cluster is a group of interlinked pages that cover one topic in depth: a central pillar page plus several satellite pages on specific subtopics, all linking to each other. Search engines read the group as a single authoritative resource, which helps every page rank better than it would on its own.
Why do topic clusters rank better than single posts?
Topic clusters concentrate authority. When many interlinked pages cover one subject, search engines treat the site as a credible source on it, so the whole group lifts together. In the case above, building roughly 24 linked pages took a topic from about 14 to 1,174 organic clicks per 28 days, far more than one post would earn.
How many pages does a content cluster need?
There is no fixed number, but depth is the point. A useful cluster pairs one pillar page with enough satellite pages to cover the real questions people search, often a dozen or more. The speeches cluster above reached about 24 interlinked pages. Add pages as long as each one answers a distinct, searched question.
How long does a content cluster take to rank?
Expect a slow start, then a sharp climb. Rankings lag publishing because search engines need time to crawl, index, and trust new pages. The cluster above sat nearly flat for ten months, then grew about 80 times over roughly thirteen weeks once the pages were built and interlinked. Build ahead of the demand you want to catch.
What is the difference between a pillar page and a cluster page?
A pillar page gives the broad overview of a topic and links out to every subtopic. Cluster pages go deep on one specific question each and link back to the pillar. The pillar earns breadth, the satellites earn long-tail terms, and the internal links between them pass authority in both directions.
How does internal linking help a cluster rank?
Internal links pass authority between related pages and signal that the pages belong together. That shared authority is why a late-added page can rank quickly: in the case above, the final pillar reached page one within weeks because the surrounding cluster had already earned trust. Link every cluster page to the others, not only to the pillar.
The takeaway in one line: stop shipping posts and start building topics, because depth and internal links compound in a way that single articles never do.
Want to know which of your own quiet pages are sitting on unmet demand? That is the first thing a free diagnosis looks for.
Continue Reading:
More From This Series
- Clusters Compound. Scattered Posts Plateau.
- The CTR Fix: Same Ranking, More Clicks
- From Zero: A Cold-Start Cluster
- The Experiment We Killed
More from TDM Insights
- Topic Clusters and Pillar Pages: How They Work
- Internal Linking: The Most Underrated SEO Lever
- How Many Blog Posts Do You Need to Rank?
- Content Strategy for Founders: A Library That Ranks, Gets Cited, and Pays for Itself
- How Long Does SEO Take? An Honest Timeline
Explore TDM Insights Topics