A topical map is the page-level plan for owning a subject: every page you need, what each one targets, and how they link together. Without one, content production turns into random publishing, and random publishing produces pages that cannibalize each other while leaving obvious subtopics uncovered.
Building one takes five steps. None of them require special tools, though tools compress the timeline; a careful manual build takes a couple of days of research for a mid-sized topic.
Step 1: pick the central entity
Start with the thing you want to be known for, phrased as an entity rather than a keyword. “Payroll software” is an entity; “best payroll software 2026” is one of hundreds of queries about it.
The distinction matters because search engines evaluate sites at the topic level. If your site covers payroll software thoroughly (setup, compliance, pricing, integrations, comparisons, mistakes), you become a credible result for queries you never explicitly targeted.
Pick one central entity per map. If your business spans several (payroll, invoicing, expense tracking), build one map per entity rather than a mega-map, so each cluster stays coherent.
Step 2: extract the entities around it
Now list everything a genuine expert would cover. Four sources, in the order I’d use them:
- The live SERP. Search your central entity and harvest subtopics from the top 10 pages’ headings, People Also Ask, and related searches.
- Competitor sitemaps. A competitor who already ranks has done part of this work; their URL structure shows how they carved up the topic.
- Entity extraction tooling. Agility Writer’s Entity Extraction tool analyzes content and ranks the entities in it, and can compare your page against a competitor URL to show which entities you’re missing.
- Your own customer questions. Support tickets and sales calls surface subtopics the SERP hasn’t saturated yet, which is where the easy rankings are.
For “payroll software,” this produces entities like tax filing deadlines, contractor vs employee classification, direct deposit setup, payroll compliance penalties, integration with accounting tools, and migration between providers.
Step 3: choose the cluster shape
Group the entities into clusters, then decide how the clusters connect. Two patterns cover almost every site:
| Pattern | Best for | Linking style |
|---|---|---|
| Hub and spoke | New sites or new topics establishing authority | A pillar page links down to 5-15 support pages; every support page links back up |
| Mesh | Large, mature sites with deep archives | Related pages cross-link on shared entities, without strict hierarchy |
Start with hub and spoke. It is easier to brief writers on, easier to validate, and easier for search engines to interpret. Meshes emerge naturally later as clusters start touching.
Step 4: turn clusters into a page list
Assign each entity a page, a page type, and a working URL. The page type must match what actually ranks: search each target query and look at the top three results. If they are all how-to guides, a product page will not rank there no matter how good it is.
The output of this step is a flat table your writers can execute: URL, target query, page type, parent hub, and the 2-3 internal links each page must carry. That table is the topical map.
Step 5: validate before you write anything
Every hour spent validating the map saves many hours of rewriting. Four checks:
- Cannibalization: no two planned URLs target the same primary intent. If two rows in your table would satisfy the same searcher, merge them.
- Depth: each cluster has enough support pages to demonstrate expertise (see the FAQ below for ranges). One hub with two spokes convinces nobody.
- Intent match: every page type matches its SERP, per step 4.
- Link paths: every planned page connects to the hub, and the hub connects to at least one existing page with authority, so the new cluster isn’t an island.
If you already have published content, run the Topical Authority Audit Tool first. It maps your current site structure, finds pages fighting over the same keywords, and outputs a create/optimize/redirect plan, which tells you how much of the new map you can build from existing pages instead of new ones.
The faster version
The manual process above is exactly what the Topical Map Helper automates: give it a keyword or just your URL (it works out the main topic), and it produces the cluster structure and page list, in any of the languages Agility Writer writes. For local businesses it also generates the location-page structure.
One honest caveat, and it applies to every mapping tool on the market: no automated tool produces a 100% perfect topical map. Treat the output as a strong first draft, then run step 5 on it yourself before you brief a single article. Review and refine is part of the workflow, not a failure of it.
What to do next
Pick your most important topic and run steps 1-2 this week; that’s the research half, and it’s reusable even if your cluster shape changes later. Then either finish the map manually or run it through the Topical Map Helper and spend your time on validation instead of spreadsheet assembly. For the tool landscape beyond our own, see our review of topical map tools.