Table of Contents
- What Keyword Mapping Actually Decides
- Step 1: Group Keywords Into Intent Clusters, Not a Raw List
- Step 2: Identify the Intent Behind Each Cluster
- Step 3: Match Intent to the Page Type Google Already Rewards
- Step 4: Check Whether an Existing Page Already Owns the Cluster
- Step 5: Apply the “Does This Deserve Its Own Page” Test
- Step 6: Assign, Prioritize, and Record the Decision
- Where Mapping Breaks: Cannibalization and Mismatched Page Types
- A Worked Example
- How This Fits Into a Real SEO Build
- FAQ
Keyword research tells you what people search for. It does not tell you which page should own each term. That second decision, matching a keyword cluster to a specific page, is what actually determines whether your site ranks cleanly or fights itself for the same query.
Most teams treat mapping as a formatting step: paste keywords into a spreadsheet next to a list of URLs.
That is not mapping.
Mapping is a decision.
For every keyword cluster, you are deciding three things at once: which page type Google expects for that search, whether an existing page can own it, and whether it deserves a page of its own at all.
Get that decision wrong, and you end up with two of the same page competing for one ranking, or a page trying to answer three unrelated questions at once.
This guide walks through the framework we use to make that decision, cluster by cluster, before we build or rewrite a single page.
What Keyword Mapping Actually Decides
A keyword map is not a list. It is a set of assignments, and each assignment answers four questions:
- Which cluster. A group of keywords that share the same searcher intent, not just similar wording.
- Which page type. Service page, category page, location page, comparison page, blog article, or another format Google already rewards for that query.
- Which URL. An existing page that should absorb the cluster, or a new page that does not exist yet.
- What priority. Whether the cluster is worth building now, worth queuing, or not worth a dedicated page at all.
Skip any one of those four and the map turns into a wish list instead of a build plan.
A spreadsheet with 300 keywords and no page assignment tells your team nothing about what to build first, and a page assignment with no priority tells them nothing about what to build first either.
Step 1: Group Keywords Into Intent Clusters, Not a Raw List
Before you can map anything, group your keywords by what the searcher is trying to accomplish, not by how similar the words look.
“Commercial cleaning cost” and “commercial cleaning services” look alike, but the first belongs to a pricing question and the second belongs to a service page. Group by task first. Wording overlap is a secondary signal, useful for confirming a cluster, not for defining one.
A cluster is ready to map when every keyword inside it would be satisfied by the same page.
If two keywords in a cluster would need different content to answer them properly, split the cluster before you go any further.
Step 2: Identify the Intent Behind Each Cluster
Every cluster carries a dominant search intent. Naming it correctly is what makes the rest of the mapping decision possible.
- Commercial intent. The searcher is comparing options or evaluating a provider, but has not decided to buy yet.
- Transactional intent. The searcher is ready to act: book, request a quote, sign up, or buy.
- Informational intent. The searcher wants an explanation, a definition, or a how-to answer.
- Navigational intent. The searcher is looking for a specific brand, tool, or known destination.
- Local intent. The searcher wants a provider that serves their city, service area, or region.
A cluster can carry more than one intent signal, but one should dominate. If you cannot name the dominant intent in a sentence, the cluster still needs work before it gets mapped.
Step 3: Match Intent to the Page Type Google Already Rewards
Once the intent is clear, check what page type is already winning for that cluster.
This is the step most keyword research skips, and it is the one that prevents the most expensive mistake: building the wrong format for the query.
| Dominant intent | Page type Google typically rewards | What breaks if you build the wrong one |
|---|---|---|
| Transactional | Service page, product page, or booking page | A blog post ranks below providers with a real offer page |
| Commercial | Comparison page, pricing page, or category page | A generic landing page loses to a page built to compare options |
| Local | Location page or service-area page | A single national page can’t hold city-level rankings |
| Informational | Blog article, guide, or glossary entry | A commercial page reads as thin and fails to satisfy the question |
| Navigational | Homepage or the specific named page | A new page competes with the page it should be reinforcing |
If your SERP research shows service pages dominating a cluster, don’t answer it with a blog post, no matter how much content you already have planned.
If category pages dominate, a thin landing page will not hold its position. The SERP is telling you the format before you write a word of copy.
Step 4: Check Whether an Existing Page Already Owns the Cluster
Before creating anything new, map the cluster against your current site. Three outcomes are possible:
- A page already covers it well. Nothing to build. Note the assignment so no one else creates a competing page later.
- A page covers it partially or poorly. The cluster belongs to that page, but the page needs to be expanded, rewritten, or repositioned to fully satisfy it.
- No page covers it. The cluster is a real gap and becomes a candidate for a new page.
This step is where most cannibalization gets prevented, not fixed.
If you skip it, teams keep creating new pages for clusters an existing page already half-answers, and now two pages compete for the same query.
Step 5: Apply the “Does This Deserve Its Own Page” Test
Not every keyword variation earns a dedicated URL.
Before creating a new page, or splitting a cluster off an existing one, check it against these five questions:
- Does the searcher’s task actually differ from the page that would otherwise absorb it?
- Does it require a different central subject, not just different wording?
- Does the SERP show a different page type or a materially different set of ranking pages?
- Is there enough commercial or search value to justify a standalone page instead of a paragraph or FAQ entry?
- Could an existing page answer this completely if it were expanded, rather than forking into a new URL?
If the answer to most of these is no, the keyword belongs inside an existing page, as a section, a paragraph, or an FAQ entry, not as its own URL. Answering yes to most of them is what justifies a new build.
Step 6: Assign, Prioritize, and Record the Decision
Once a cluster has an intent, a matched page type, and a page decision, record it. A usable keyword map shows, for every cluster:
- the cluster and its dominant intent
- the assigned page, existing or new
- the action: keep, expand, rewrite, merge, or build
- the priority, based on business value and how much of the gap is currently uncovered
This is the difference between a keyword list and a build plan. A writer, a developer, and a founder should all be able to look at the same map and know what happens next without asking a follow-up question.
Where Mapping Breaks: Cannibalization and Mismatched Page Types
Two failure patterns account for most of the mapping problems we see on existing sites.
Cannibalization happens when two or more pages end up targeting the same cluster, usually because a new page got built without checking what already existed. The fix is not writing better copy on either page. It’s consolidating them into one page that fully owns the cluster, and redirecting or repositioning the page that loses.
Mismatched page type happens when the page format doesn’t match what the SERP is already rewarding, a blog post targeting a transactional cluster, or a single national page targeting a cluster that needs city-level coverage. The fix is rebuilding the page as the correct type, not adding more words to the wrong one.
Both patterns get caught at Step 4 and Step 3 of this framework. Teams that skip straight from keyword list to content calendar are the ones that create them.
A Worked Example
A commercial cleaning company has these keywords in their research: “office cleaning services,” “office cleaning cost,” “commercial cleaning company near me,” and “how often should offices be cleaned.”
Mapped correctly:
- “Office cleaning services” and “commercial cleaning company near me” share transactional and local intent. They map to the core service page, with city or service-area pages absorbing the “near me” and location variants.
- “Office cleaning cost” is commercial intent. It gets its own section, or its own page if the search volume and competition justify it, rather than living buried inside the service page.
- “How often should offices be cleaned” is informational. It does not belong on a commercial page at all. It maps to a blog article, which can then link back to the service page as the next step for a reader who is ready to act.
Four keywords, three different page assignments, because the intent behind each one is different even though they all relate to the same business.
How This Fits Into a Real SEO Build
This framework is the decision layer.
Turning it into a finished site is the execution layer, which is what our keyword research services actually deliver: a page map built cluster by cluster, checked against your existing site, and prioritized so your team knows what to build first.
The pattern holds up in practice.
When an IT broker’s keyword mapping and enterprise service pages were rebuilt around buyer intent instead of a scattered keyword list, organic visitors grew from roughly 1,500 a month to more than 6,000, with 120% more inbound leads.
A nearshore staffing company’s keyword architecture went from zero organic visibility to 12,500 monthly clicks once its clusters were mapped to the right pages instead of left as a keyword list with no build plan.
If you want to see how many pages your own keyword gaps actually call for, the SEO content planner is a free way to size that before committing to a full rebuild.
FAQ
What if a keyword doesn’t fit any page type on my site?
That’s a real content gap, not a mapping error. Run it through the Step 5 test. If it earns its own page, queue it. If it’s a smaller supporting question, it likely belongs as a section or FAQ entry on a page you already have.
Does every keyword variation need its own page?
No. Most variations belong inside an existing page’s cluster. A new page is only justified when the searcher’s task, the expected page type, or the SERP itself is genuinely different from what an existing page already covers.
How often should a keyword map be revisited?
Whenever you publish, merge, or significantly change priority pages, and at minimum every quarter. Search behavior shifts, competitors rebuild pages, and a map that isn’t checked against reality becomes a map of decisions that no longer apply.
Can two pages target related keywords without cannibalizing each other?
Yes, as long as each page satisfies a distinct intent. A service page and a pricing-focused page can both rank for related terms if one answers “what is this” and the other answers “what does this cost.” The risk starts when both pages try to answer the same question.
