We built 4 topic clusters from scratch. Six months later, organic sessions from targeted keywords were up 340%. Here is the spreadsheet structure and how to replicate it.
A topic cluster is a group of content pages built around one central topic. One pillar page covers the topic broadly and comprehensively. Multiple spoke pages cover specific subtopics in depth. Every spoke links back to the pillar. The pillar links out to every spoke. The internal linking structure signals to Google that your site has concentrated expertise in this topic area, not just a single lucky post.
We built four clusters for Striveloom over a six-month period in 2025. Before the clusters, our organic sessions from content pages were 480 per month. After six months with all four clusters live and indexed, they were 2,112 per month from the same page types. The improvement is attributable to the cluster architecture and not to other factors because we made no significant changes to our service pages, homepage, or link building during that period.
This is the playbook we used. The spreadsheet template is free. Here is how to build your own.
Agency websites face a specific SEO challenge: they typically cover multiple service areas (web development, SEO, paid ads, content, design) that have little topical overlap. A site that publishes one post about web development, one about SEO, one about paid ads, and one about design has four isolated pieces of content, none of which builds on the others. Google sees a site trying to be expert in everything and expert in nothing.
Topic clusters solve this by concentrating content depth in specific areas. A cluster of 8 posts on "Next.js web development for startups" tells Google something very different from 8 posts scattered across 8 different topics. The cluster signals genuine topical authority. It also creates internal linking patterns that distribute authority from the pillar page to related spoke pages, helping each spoke rank better than it would as a standalone post.
HubSpot popularized the topic cluster concept in 2017 based on research into how Google's Hummingbird and subsequent updates rewarded topic expertise over keyword density (per HubSpot, 2017). The logic has only become more applicable as Google has shifted toward evaluating E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) as a quality signal.
For agencies specifically, clusters solve another problem: they force focus. If you decide to build a cluster around "AI-powered marketing automation," you have to write 6 to 8 posts about that specific topic, which requires genuine expertise. Agencies with genuine expertise in a specific area produce better clusters than agencies trying to cover everything.
Our clusters follow a consistent architecture. Each cluster has one pillar post, five to seven spoke posts, and two to three FAQ posts. Here is the template:
The pillar page is comprehensive but intentionally shallow on subtopics. If the pillar is "Complete Guide to Next.js Web Development for Startups," it covers architecture decisions, build configuration, deployment options, and performance optimization at a high level. It does not go deep on any one of them. Each of those subtopics gets its own spoke post where depth lives.
This is the mistake most agencies make when building clusters: the pillar tries to cover everything in depth, leaving no room for spoke posts, and the spoke posts end up duplicating pillar content. The pillar should be 30% broad coverage of all subtopics and 70% structure (linking, navigation, definitions). The spoke should be 100% depth on one specific subtopic.
For an agency, pillar topics should match your service areas. If you offer SEO, web development, and content strategy, start with one cluster per service area. Pick the service area where you have the most genuine expertise first.
Good pillar topic: "SEO for B2B SaaS Companies" Bad pillar topic: "Digital Marketing" (too broad, not achievable topical authority)
Use a keyword research tool to find questions and subtopics that fall under your pillar topic. You want keywords that are:
For "SEO for B2B SaaS Companies," example spokes might be:
This is where most teams skip a step. Map the internal links before writing any content. For every spoke post, define: which pillar page does it link back to? Which other spokes does it link to (maximum 2, ideally 1)? Where in the post does the internal link appear (not in the sidebar, not in the footer, in the body text where it reads naturally)?
Our cluster spreadsheet has a column for each of these. Before a spoke post goes live, the internal link map is complete. This prevents the common failure mode where posts are written and published without internal links, requiring a separate audit pass to add them later.
The pillar comes first because it defines the structure of the cluster. Writing spoke posts before the pillar often results in spokes that are too broad (duplicating what the pillar should cover) or too narrow (missing the connection to the pillar topic). Write the pillar, then write the spokes to fill the gaps the pillar intentionally left.
Publish the pillar first. Then publish spokes one at a time, each linking back to the pillar. After each spoke is published, add a link to that spoke from the pillar. The pillar should be a living document that grows its internal link count as spokes are published.
The mistakes we made on our first cluster and fixed on subsequent ones:
Spokes that compete with the pillar. If the pillar targets "agency content strategy" and a spoke targets "content strategy for agencies," they are competing for the same searchers. Make sure each spoke targets a clearly different, more specific query than the pillar.
No FAQ posts. FAQ posts target question-format queries that pillar and spoke posts miss. "What is the difference between a pillar page and a spoke page?" is a specific enough question to deserve its own 800-word FAQ post that links to the pillar. FAQ posts are fast to produce and often capture high-intent searches.
Internal links buried in footers or sidebars. Google discounts links in boilerplate locations. Internal links for cluster architecture should appear in the body text of the post, where they are contextually relevant. A link that reads "For a deeper look at LCP optimization, see our guide to [Next.js performance]" in the body of a technical post is worth more than the same link in a "related posts" widget at the bottom of the page.
Clusters without a publication schedule. Pillar posts are large. Spoke posts take time. Without a committed publication schedule, clusters stall at 2 or 3 posts and never reach the critical mass where topical authority kicks in. We committed to publishing one spoke post per week per cluster and treated that as a non-negotiable production commitment for six months.
Four clusters. Eight posts each. 32 total pieces of content published over six months.
Organic sessions from cluster-targeted keywords: 480 per month before clusters, 2,112 per month six months after cluster launch. That is a 340% increase from the content initiative alone. Our domain's total organic sessions also grew during this period, which suggests the clusters contributed to overall domain authority as well as direct ranking improvements.
Three of the four pillar pages now rank on page 1 for their primary keyword. The fourth ranks in position 14. The spoke posts average position 9 across all targeted keywords.
Pick one service area. Write one comprehensive pillar post. Write four spoke posts that each target a specific subtopic your pillar covers briefly. Link them all together before publishing. Publish on a schedule you can actually maintain.
That is the minimum viable cluster. It is not glamorous. It is the content architecture that produces compounding search results over time, and it is the same architecture that we and many other agencies have used to build sustainable organic traffic.
If you want the spreadsheet template we use, it is available at Striveloom's resources page. It has the pillar, spoke, FAQ, and case study tabs with the column structure described above, pre-populated with our web development cluster as an example you can modify.
Topical authority is built slowly, one cluster at a time. But it compounds. And it is far more defensible than any single blog post that happens to rank well this month.
A topic cluster is a group of related content pages built around one central topic. A pillar page covers the topic broadly and links to multiple spoke pages, each of which covers a specific subtopic in depth. Every spoke links back to the pillar. The internal linking structure signals topical authority to Google and distributes link equity across the cluster. HubSpot popularized this model in 2017, and it remains one of the most consistent organic traffic strategies for content-heavy sites.
A minimum viable cluster has one pillar page and four spoke posts, for a total of five pieces of content. This is enough to begin signaling topical authority. Clusters typically see stronger results as they grow: 8 to 10 posts covering a topic thoroughly tend to outperform 4 to 5 posts covering the same topic partially. The exact number depends on how competitive the topic is and how much search demand exists across the subtopics. More competitive topics require more depth to establish authority.
A pillar page is broader and longer than a typical blog post (usually 2500 to 3500 words), covers the full scope of a topic at a medium level of depth, and explicitly links to spoke posts for deeper coverage of specific subtopics. A regular blog post targets one specific query and covers it thoroughly without linking to related subtopic posts. The pillar's job is to be the authoritative central document for a topic. The spokes' job is to answer specific questions the pillar intentionally leaves for dedicated pages.
Yes. Internal links from pages with higher authority pass ranking signals to the pages they link to. A pillar page that has earned inbound links from external sites passes a portion of that authority to spoke pages through internal links. This is why publishing the pillar first (and earning links to it) before publishing spokes can accelerate spoke rankings. It is also why the pillar should be linked from the homepage and other high-authority pages on the site — increasing the pillar's authority increases what it can pass to spokes.
A landing page is conversion-focused: it exists to get visitors to take a specific action (book a call, sign up, purchase). A pillar page is information-focused: it exists to answer questions comprehensively and demonstrate topical expertise. Some agencies combine both functions in a single page, which is a valid strategy if done carefully. But purely educational pillar pages tend to earn more external links (because they are linkable resources rather than sales tools) and rank better for informational queries, which often represent the majority of search demand in an agency's topic area.
In our experience, new clusters start showing measurable ranking movement for spoke posts within 60 to 90 days of publication. Pillar pages for competitive topics typically take 4 to 6 months to reach page 1. Clusters built in less competitive niches can rank faster. The compounding effect — where ranking spokes earn backlinks that benefit the pillar, which in turn benefits all spokes — typically becomes visible around month 4 to 6. This is slower than paid search but produces more durable traffic that does not disappear when a budget is cut.
Founder & CEO of Striveloom. Software engineer and Harvard graduate student researching software engineering, e-commerce platforms, and customer experience. Builds the agency that ships like software — one team, one pipeline, one platform. Writes on AI agencies, web development, paid advertising, and conversion optimization.
Everything a startup founder needs to know about building a website in 2026 — tech stack choices, realistic budgets, timelines, and the mistakes that kill product launches.
Three posts on our blog earned 9 AI Overview citations in 90 days. Here is what they had in common, what we changed, and what did not work at all.
We ran a 12-point SEO audit on 50 digital agency websites. 47 had critical failures. Here is the full dataset and what the top three did differently.
Book a free 30-minute call to scope your project. Fixed pricing, transparent timelines.
| Content Type | Target Length | Internal Links | Primary Goal |
|---|
| Pillar page | 2500-3500 words | Links to all spokes | Rank for broad topic keyword |
| Spoke post | 1500-2000 words | Links to pillar + 1-2 related spokes | Rank for specific subtopic keyword |
| FAQ post | 800-1200 words | Links to pillar | Capture question-format searches |
| Case study | 1000-1500 words | Links to relevant pillar | Provide E-E-A-T evidence |
