Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Templates

Keyword Map Template

A keyword map template: keyword cluster, primary query, intent, owner URL, page type, current state, action, priority and the brief or refresh it feeds.
Templateintermediate10 min readUpdated

What this template is for

Use the map for any site that publishes more than a handful of pages for search: a shop with categories, products and guides, a service business with location and service pages, a publisher with years of articles. It is most valuable at three moments. When a site is planned or rebuilt, it decides which pages should exist. When rankings are disappointing, it exposes pages competing for the same queries and clusters nobody owns. And before any new piece of content is commissioned, it answers the question every proposal should face first: does a page already own this?

The map sits between documents that already exist on this site. Before it, the clustering guide turns a raw keyword list into clusters with a head term, and the intent classification guide labels each one. If an AI tool drafted those clusters or labels, the AI for keyword research guide lists the checks against real query and results data to run before they become rows. After it, every row marked create or improve becomes a content brief, and the SEO content brief tutorial adds the coverage and metadata a ranking page needs. Around it, the topic clusters model groups the owner pages into hubs and spokes; the keyword map decides the pages, the topic cluster decides how they link. When the map reveals pages competing for one query, the cannibalisation guide holds the decision rules and the execution sequence.

The map is a working document, not a report. It is opened whenever a page is proposed, merged, retired or rebuilt, and its history explains why each page on the site exists.

The template

Build the map as a spreadsheet with one row per cluster and these columns. Add a status column and a last-reviewed date so the map can be maintained, and keep member queries in a linked tab rather than in the row, where they swamp the decision.

FieldWhat to write
ClusterA short cluster name or ID and its member queries, held in the linked tab
Primary queryThe head query that names the cluster: the one with most demand whose results page best represents it
IntentInformational, commercial, transactional, navigational or local, read from the results page, and the format that ranks
Owner URLExactly one URL, existing or proposed, that owns the cluster
Page typeGuide, category, product, tool, landing page or location page, matching the format the results page rewards
Current stateRanks well, ranks poorly, wrong page ranking, no page, or several pages competing, with the competing URLs
ActionKeep, improve, consolidate, split or create, with the URLs any redirect involves
PriorityThe order of work and the reason, from business relevance, demand, gap and effort
Brief and linksThe brief or refresh ticket the row feeds, and the internal links to add, with their anchor text

How to fill each section

Cluster

Each row starts from a cluster built in the clustering guide: queries that the results page treats as the same request, because mostly the same pages rank for them. Name the cluster for humans and give it a stable ID, because names get edited and IDs are what the brief and the status tracker refer to. Keep member queries in a separate tab with their estimated Search Volume; the row only needs the count and the link. If two clusters turn out to share their results, merge them here rather than mapping both.

Primary query

Pick the query with the most demand whose results page best represents the cluster. It names the row and drives the page's title, but it is not a phrase to repeat: the variants are covered naturally by a page that answers the cluster well, which is what Google's SEO starter guide, listed in the sources, describes when it advises writing for readers rather than for keyword lists. Where demand needs translating into a rough traffic expectation, the SEO traffic estimator gives a range for planning; treat it as a planning aid, not a forecast.

Intent

Read the intent from the results page, not from the words. Search the primary query, look at what ranks — guides, category pages, product pages, tools, local results — and record the Search Intent with the format that wins. The intent classification guide sets out the categories and the modifier rules. Intent is the hard boundary of the map: queries with different intents never share a row, however similar the words.

Owner URL

Write exactly one URL. For an existing page, use its canonical URL. For a page that does not exist yet, write the proposed URL and check it against the site architecture so it sits in the right section. One URL may own several clusters only when they share intent and the page genuinely serves them all; one cluster may never have two owners. This column is what the keyword mapping tutorial calls absolute, and it is the column everything else in the map relies on.

Page type

Match the page type to the format the results page rewards. A how-to cluster needs a guide; a "buy" cluster needs a category or product page; a calculation needs a tool. The search intent guide explains why format, not wording, decides whether a page can compete. A mismatch here is the most common reason a page with good content does not rank, and it changes the action: the fix is a new page of the right type, not more words on the wrong one.

Current state

Pull the site's ranking pages for each cluster from Search Console's performance report, filtered by query, and classify the row: ranks well, ranks poorly, wrong page ranking, no page, or several pages competing. For competing pages, list every URL with impressions for the cluster, because the cannibalisation guide treats this as evidence, and similar titles alone are only suspicion.

Action

Each state leads to an action. Keep a page that fits and ranks. Improve a page that fits but ranks poorly. Consolidate competing pages into the owner and redirect the rest. Split a page stretched across clusters that deserve their own pages. Create a page where none exists or where the existing one is the wrong type. For consolidations, write the redirect pairs in the row; Google's guidance on consolidating duplicate URLs, listed in the sources, explains how redirects and canonical tags each indicate the preferred URL: a redirect suits a page being retired, a canonical tag a duplicate that must stay live. For a split, the tutorial on splitting a page that serves different intents works through dividing the live page.

Priority

Order the rows by relevance to the business's current priorities, demand, the gap between current and achievable position, the strength of the competition and the effort of the action. Write the reason in a few words, because a priority without a reason is re-argued every planning meeting. Structural fixes — consolidations and splits — usually come before new content, because new pages written into an unresolved overlap inherit the problem.

Link the brief or ticket the row feeds, so the writer starts from the row's purpose. Then list the internal links to add: which pages should link to the owner URL, with which anchor. The internal linking guide explains anchor discipline: anchors for a cluster's primary query point at its owner and nowhere else, which is the map applied to the links.

Filled example

Below, a section of the map is completed for an invented business. The URLs, positions and figures are illustrative.

Illustrative example — not a real result

Tamberly Kitchenware: pans, knives and bakeware clusters

ClusterPrimary queryIntentOwner URLPage typeCurrent stateActionPriorityBrief and links
PAN-01 non-stick frying pans (14 queries, about 9,900 searches a month)non stick frying panTransactional; category grids rank/pans/frying-pans/non-stickCategorySeveral pages competing: the category page (position 14) and a 2023 blog post "Our favourite non-stick pans" (position 19) swap for the primary queryKeep the category as owner; rework the blog post as the buying guide for PAN-02 and remove its "buy" phrasing1: high revenue, two pages splitting signals, cheap fixRefresh ticket SEO-112. Link from homepage pans tile and from PAN-02 guide with anchor "non-stick frying pans"
PAN-02 choosing a frying pan (9 queries, about 2,400 a month)best frying pan materialCommercial; buying guides rank/guides/choosing-a-frying-panGuideWrong page: the category page ranks at position 31Create from the reworked 2023 post; 301 the post's old URL to the new guide2: supports PAN-01 and three other pan categoriesBrief CB-041. Link to each pan category with its category name as anchor
KNF-01 chef's knives (11 queries, about 6,600 a month)chef knifeTransactional; category grids rank/knives/chefs-knivesCategoryRanks poorly: position 22, thin category copy, no filters for blade lengthImprove: add blade-length and steel filters and a short buying note above the grid3: high demand, larger effortTicket SEO-118 with the merchandising team
KNF-02 knife sharpening (17 queries, about 4,400 a month)how to sharpen a kitchen knifeInformational; how-to guides and videos rank/guides/how-to-sharpen-a-knifeGuideSeveral pages competing: a guide (position 8) and an FAQ answer (position 27)Consolidate: merge the FAQ answer into the guide, 301 the FAQ anchor page, add a short video4: ranks already; protects positionRefresh ticket SEO-120. Link from KNF-01 and from each whetstone product page with anchor "how to sharpen a knife"
BAK-01 loaf tins (6 queries, about 1,900 a month)loaf tinTransactional; category grids rank/bakeware/loaf-tinsCategoryRanks well: position 4Keep6: monitor onlyNo action; already linked from bakeware hub
BAK-02 sourdough proving baskets (8 queries, about 1,300 a month)bannetonTransactional and commercial mixed; products and guides both rank/bakeware/proving-basketsCategoryNo page: the products sit in a general "baking accessories" categoryCreate a category, with a short "which size" section for the commercial half of the results5: modest demand, new range launchingBrief CB-044. Link from bakeware hub and from the sourdough recipe page with anchor "proving baskets"

Three months later, PAN-01's category page held position 6 alone for its primary query and the reworked PAN-02 guide ranked at 11; KNF-02 was unchanged at position 8 after the merge.

Common mistakes

  • Volumes without decisions. A spreadsheet of queries and demand is research, not a map. Every row needs an owner URL and an action.
  • Two owners for one cluster. The problem the map exists to prevent. If two URLs appear in the owner column, the row is not finished.
  • Intent guessed from the words. Similar phrases can ask for different things. Read the results page and record the format that ranks.
  • The wrong page type. A category page cannot own a how-to cluster, and a guide rarely ranks for a "buy" query. Change the page, not the word count.
  • New content before structural fixes. Writing into an unresolved overlap creates a third competitor. Consolidate and split first.
  • Canonical tags where redirects belong. A retired page should redirect to its owner; a canonical tag is a hint for pages that must stay live.
  • Internal links ignored. Anchors pointing a cluster's primary query at several pages undo the map. Record the links each row needs.
  • A map built once. Results pages merge and split, and new pages appear. Review the map on every publish, merge or retirement and at a regular review, and date each row.

The content brief that each create or improve row feeds is in the templates library, with the other planning documents.

Terms used on this page

Tools to try

Frameworks

Where this page is used

Also related

Sources

  1. SEO Starter Guide — Google Search Central
  2. Consolidate duplicate URLs — Google Search Central

Related