A focus keyword is the one search query you decide a page should rank for. You pick it before you write, and it shapes the title, the URL, the opening and the structure of the page. The common mistake is treating it as something Google reads. Google never sees the box in your SEO plugin; it sees the page. This guide shows how to choose well: a five-test framework, a worked example, a placement cheat sheet, and the pages where setting one makes things worse.
What a Focus Keyword Is (and What Google Never Sees)
The term comes from WordPress SEO plugins. Yoast calls the field “focus keyphrase” and defines it as “the search term you want a page or post to rank for.” Rank Math and All in One SEO have the same field. When you type a phrase in, the plugin checks your draft against it: is it in the SEO title, the slug, the first paragraph, the subheadings? Then it shows a coloured score.
The field’s value is not published as a tag that search engines rank on. The closest thing that ever existed was the keywords meta tag, and Google said in 2009 that it “doesn’t use the keywords meta tag in our web search ranking.” Google’s SEO Starter Guide still repeats that today. (We cover the history in our guide to meta keywords.) So the field records a planning decision you make for yourself, not a signal you send to Google. Its value comes from choosing the right query.
Focus Keyphrase vs Primary Keyword: Same Idea, Different Tools
These terms all mean the same thing. They come from different tools:
- Focus keyphrase is Yoast’s term, a nod to the fact that most targets are several words long.
- Focus keyword is what Rank Math and most blog posts use.
- Primary keyword is the term in keyword research and keyword mapping: the one main query a page targets, as opposed to the secondary keywords it should also rank for.
There is one real difference. “Primary keyword” belongs to a site-wide keyword map, while the plugin field belongs to a single page. They stay in sync only if the map comes first. When people fill in the plugin field page by page, two posts often end up targeting the same query, and that is how most keyword cannibalization starts.
The Five-Test Framework for Choosing One
Most guides stop at “good volume, low competition.” That’s incomplete. A candidate should pass all five of these tests, in this order:
- Demand. Do real people search this phrase, and is it worded the way they search it? Use measured volume, not a guess.
- Intent fit. Do the current results show the same type of page you plan to publish? If the results are all product pages and you’re writing a guide, you’ll lose no matter how good the guide is. Our explainer on search intent shows how to read this.
- Cluster head. This is the test most people skip. Your page will rank for a group of related queries, not one. Your target should be the head of that group: the clearest, most-searched way to phrase the topic the whole group shares. Pick a variant and you optimise for the smaller version of your own topic.
- Winnability. Can your site realistically reach page one? Compare your authority with the weakest page currently ranking, not the strongest. A difficulty score helps sort candidates; the results page tells you more.
- Uniqueness on your site. No other page on your domain should already target this query. Check before you write.
Keyword tools handle tests one and four. Tests two, three and five need you to look at real results and your own site, which is where most bad choices get made.
Worked Example: Choosing the Target for One Page
Say you’re writing an HR guide and you have four candidates: “employee onboarding checklist,” “new hire checklist,” “onboarding checklist” and “employee onboarding process.”
Step 1, demand. Pull volumes for all four from one data source so they’re comparable.
Step 2, check the results for overlap. Search each phrase and compare the top ten results. Here’s the rule of thumb we use: if two phrases share roughly half or more of their ranking URLs, treat them as one topic and target both with one page. In this example the three “checklist” phrases return mostly the same downloadable checklists and templates. “Employee onboarding process” returns explainers about stages and timelines. That’s a different intent, so it needs a separate page.
Step 3, choose the head. Among the three checklist phrases, pick the one with the strongest demand that also reads naturally in a title. That becomes the primary target. The other two become secondary keywords, and you work them into subheadings and copy where they fit. You don’t build separate pages for them.
Step 4, check your own site. If an old post on new hire paperwork exists, decide now whether to merge, redirect or re-angle it.
You can do the grouping for free with our Keyword Grouping Tool. Inside SEO Rocket, the keyword research step pulls real Ahrefs volumes, difficulty and intent for every candidate, then groups them, so the cluster head is easy to spot instead of something you work out by hand.
Where to Use It, and What Each Placement Is Worth
Plugins give every placement the same green tick, but they don’t matter equally. Our ranking, based on what Google documents:
| Placement | Why it matters | Priority |
|---|---|---|
| Title tag | It’s the main source of the title link people click in results. The query matching it helps relevance and click-through. | High |
| H1 and opening paragraph | It confirms straight away that the page answers the query, for readers and for snippet selection. | High |
| Subheadings (some, not all) | Structure helps readers and engines map sections to sub-questions. | Medium |
| Meta description | It doesn’t affect ranking, but search terms are often bolded in snippets, which can lift click-through. | Medium |
| URL slug | A short, readable slug helps users. Google’s Starter Guide says keywords in the domain or URL path “alone have hardly any effect.” | Low |
| Image alt text | Use it only where it accurately describes the image. Alt text exists for accessibility first. | Low |
| Repetition in the body | Google’s spam policies call out “repeating the same words or phrases so often that it sounds unnatural.” | Avoid forcing it |
Use the keyword where a reader would expect it and nowhere else. There is no target density.
How Yoast, Rank Math and AIOSEO Handle It
The three main plugins treat the field differently:
- Yoast SEO. One focus keyphrase per page in the free version. Its assessments include “Keyphrase in introduction,” “Keyphrase in title,” “Keyphrase in slug,” “Keyphrase density” and “Previously used keyphrase.” Premium adds related keyphrases, synonyms and word forms, plus a “Keyphrase distribution” check.
- Rank Math. Rank Math’s own documentation says the free plugin allows up to five keywords per post and PRO removes the limit. The first keyword you enter is the primary one. Tests for the title, meta description, URL and opening content check only that first keyword.
- All in One SEO. It has a focus keyphrase plus an optional “additional keyphrases” field for secondary targets.
Every plugin treats the first phrase as the one that counts. Adding five keywords in Rank Math doesn’t make a page rank for five topics; the plugin just scores five strings. For a full comparison, see our Rank Math vs Yoast breakdown.
The Plugin-Score Trap
A green light means your text contains a string in the places the plugin checks. It says nothing about whether the page deserves to rank: whether it matches intent, whether it’s better than the current top ten, whether anyone links to it. Twisting a sentence to fit “best dentist Singapore cheap” into the introduction because the plugin wants an exact match reads worse and earns nothing.
Two plugin checks are still worth taking seriously. The first is Yoast’s “Previously used keyphrase” (and Rank Math’s equivalent uniqueness test), because it’s a free early warning for cannibalization. Back it up with a site-wide check in our Keyword Cannibalization Checker. The second is the title check, because the title is the placement that matters most. Ignore the rest when they hurt the writing. Google’s Starter Guide says there’s “no magical word count target” and that heading order “doesn’t matter” to Search.
When to Leave the Field Empty
Yoast’s own guide admits that “not every page needs a focus keyword.” Here’s where we’d leave the field empty, or fill it only to stop the warnings:
- Utility pages. Contact, login, cart, thank-you, privacy and terms pages. Nobody needs them to rank for a non-branded query, and many should carry a noindex anyway.
- The homepage. It usually ranks for your brand and, for a small business, maybe one core service phrase. Forcing a single keyword onto it often weakens the brand message. Our homepage SEO guide covers what to do instead.
- Pages that serve several intents at once. Hubs, directories and resource libraries rank through the pages they link to. Give the hub a broad topic label and put specific targets on the child pages.
- Pages for queries with no measurable volume. Sales enablement pages, niche B2B use cases and support articles exist because customers need them. Write them clearly and don’t chase a plugin score.
- A second page for a query you already target. If the honest target is a query an existing page already owns, don’t publish the new page. Improve the existing one.
When to Change Your Target After Publishing
Your chosen keyword is a guess made before launch. After launch, Google Search Console’s Performance report shows the queries the page actually gets impressions for. Three patterns are worth acting on:
- The page ranks for a different head term. If most impressions come from a phrase you treated as secondary, and its intent matches the page, retitle the page around that phrase. Google already sees the page that way.
- Two of your URLs swap places for the same query. That’s cannibalization. Merge them, or rewrite one for a different intent, then 301 the loser.
- Impressions but no clicks on the target. The page is relevant but the snippet doesn’t win the click. Rewrite the title and description before touching the body.
SEO Rocket’s rank tracking and Search Console integration show this side by side, so you can see when a page’s real query set has moved away from what you planned. Its article writer’s validation gates enforce title and meta-description limits before anything ships. It’s the same discipline behind a playbook proven across 1,000,000+ ranking pages: decide the query, check it against the live results, then let the data correct you.
FAQ
Should every page have exactly one primary keyword?
Every page you want to rank should have one primary target, plus the secondary keywords that share its search results. Utility pages and hubs often don’t need one at all.
Does Google use the Yoast focus keyword?
No. It’s a field inside the plugin that drives Yoast’s content checks. Google ranks the published page, not the plugin settings, and it confirmed long ago that it ignores the keywords meta tag.
Can two pages target the same keyphrase?
They shouldn’t, if both are meant to rank for it. The two pages compete for the same query, and Google may swap between them or rank neither well. Give each page its own query, or merge them.
Should the keyphrase be an exact match everywhere?
No. Use the exact phrase in the title and early in the page where it reads naturally. Elsewhere, use variants and plain language. Google matches meaning, not just strings.
The Short Version
Choose the focus keyword before you write, and choose it from the search results, not the plugin. It should have real demand, match the intent of what already ranks, sit at the head of its cluster, be winnable for your site, and belong to no other page on your domain. Put it where readers look first: the title, the H1 and the opening. Then stop. For the research behind each test, start with our guide on how to do keyword research, and revisit the choice once Search Console shows you what the page actually ranks for.
