Introduction

This guide is documentation-based research, not a live API test. It is not a Frase trial, and no scores, rankings, or outcomes were measured here. This article may contain affiliate links. For related background, see /frase-review/.

Frase’s Content Scores page states that keywords only count toward the SEO score when they appear as a whole word; a target term tucked inside a longer word will not earn credit [https://docs.frase.io/feature-reference/content-scores-explained.md]. That single rule explains a common troubleshooting pattern: the overall score lags, EEAT and GEO look acceptable, yet the draft seems to discuss the tracked topic. The SEO score may simply be blind to embedded terms.

The reader problem is diagnostic, not editorial. You need to separate two cases: a tracked keyword that is genuinely absent, and one that is present only inside a longer word. The first may justify new coverage. The second may only need a phrase rewritten. Frase’s documentation supplies the rule; the workflow below is our suggested process, not a vendor API contract. The vendor documents a score range and general components, but it does not document how many points any single keyword contributes, so score movement after a rewrite is not predictable from the documentation alone.

Documented Workflow

Frase documents that scores appear in the top-right corner of the Editor in any article, and that you can click any score to see what is helping or hurting it [https://docs.frase.io/feature-reference/content-scores-explained.md]. The Score tab in the Editor’s right side panel shows an Overall Score across EEAT, GEO, and SEO, and each score can be expanded to see contributing components and gaps [https://docs.frase.io/core-workflows/optimize-existing-content.md]. The Checklist tab includes topic coverage, where you can manage primary and secondary keywords, add and edit queries, set keyword priority, and track coverage [https://docs.frase.io/core-workflows/optimize-existing-content.md].

The documented mechanics relevant to this troubleshooting path are narrow: whole-word matching is the scoring rule, and the expanded score view is where coverage gaps are surfaced. The documentation also notes that suggestions can be applied with one click or manually, and that some improvements need your input and cannot be automated [https://docs.frase.io/feature-reference/content-scores-explained.md]. It does not document a dedicated whole-word diagnostic, a per-keyword point value, or a guaranteed score change after a rewrite. Treat those as unknowns rather than assumptions.

Our suggested process is to use the expanded SEO score as a gap list, then audit the draft copy against that list before changing anything. The vendor contract describes where suggestions live and how they can be applied. Our process describes how to classify each flagged term so you do not rewrite a sentence that already contains the keyword in a different form.

Practical Walkthrough

This walkthrough is our original operational sequence. It uses documented Frase surfaces but does not claim undocumented behavior. It is also not a live API test.

Open the article in the Editor and locate the score display in the top-right corner, as documented [https://docs.frase.io/feature-reference/content-scores-explained.md].
Open the Score tab in the right side panel and expand the SEO score to see which components are contributing and where gaps appear [https://docs.frase.io/core-workflows/optimize-existing-content.md].
Open the Checklist tab and review topic coverage for the primary and secondary keywords you are tracking [https://docs.frase.io/core-workflows/optimize-existing-content.md].
For each tracked keyword, search the draft copy for that exact term as a standalone whole word. Do not count a longer word that merely contains the term, because the documentation says an embedded term will not earn credit [https://docs.frase.io/feature-reference/content-scores-explained.md].
Classify each tracked keyword into one of three states: standalone present, embedded only, or genuinely absent. Record the state before editing.
For an embedded-only term, decide whether rewriting the sentence to use the standalone phrase preserves meaning and readability. If it does, rewrite manually or use a documented suggestion. If it does not, note the exception and consider accepting the gap.
For a genuinely absent term, decide whether the draft needs new coverage or whether the term is out of scope. Absence is not automatically a defect; it is a coverage decision.
Re-check the expanded SEO score after the change and confirm whether the term is still listed as a coverage gap.

Hypothetically, a draft targets the keyword “content brief” but the copy only ever says “content briefing.” The editor expands the SEO score, sees the term flagged, rewrites one sentence to use “content brief” as a standalone phrase, and re-checks the score. The score may move, stay flat, or change for unrelated reasons. The documentation does not state how much any single keyword contributes, so the outcome is not predictable from the documentation. The observable check is whether the expanded SEO score no longer lists that term as a coverage gap, not whether the overall number reaches a particular value.

A reusable decision aid is a keyword whole-word check note. For each tracked term, record the term, whether it appears as a standalone whole word, whether it appears embedded only, whether it is genuinely absent, and the rewrite-or-accept decision. This note is our suggested artifact, not a Frase feature.

Completion Checks

Use these observable checks before treating the SEO score issue as resolved. They are based on documented surfaces, but the checks themselves are our process.

Each tracked keyword has been classified as standalone present, embedded only, or genuinely absent.
Each embedded-only term has either been rewritten to appear as a standalone whole word or explicitly accepted as a gap with a stated reason.
Each genuinely absent term has either received new coverage or been marked out of scope.
The expanded SEO score has been re-checked after the edits [https://docs.frase.io/core-workflows/optimize-existing-content.md].
The expanded SEO score no longer lists the rewritten term as a coverage gap, or the remaining gap has been documented as accepted.
Any one-click suggestion that was applied has been reviewed, because the documentation notes that some improvements need your input and cannot be automated [https://docs.frase.io/feature-reference/content-scores-explained.md].
The draft’s meaning and readability have been preserved after the rewrite; a standalone keyword that damages the sentence is not a completion.

These checks do not prove that the content will rank, be cited, or convert. They only confirm that the whole-word condition has been addressed or consciously accepted.

Limits

The documentation gives a general 70-85+ range as generally solid, but it does not give a numeric target beyond that, and it does not say how much any single keyword contributes [https://docs.frase.io/feature-reference/content-scores-explained.md]. Score movement after a rewrite is therefore not predictable from the documentation. A flat score after a correct whole-word fix is not evidence that the fix failed; it may simply mean the term carries little weight or that other components dominate.

The absence of a detection signal is not proof of human authorship, originality, or quality. Whole-word matching is a scoring rule, not a truth test. A term appearing as a standalone word does not mean the surrounding claim is accurate, useful, or well-sourced.

Vendor SEO, GEO, and EEAT scores are not measured search results and are not Google ranking factors. The documentation describes them as optimization signals and suggestions, not as guarantees. Marketing promises about visibility, citation, or authority should not be read as measured outcomes.

The documentation does not invent permissions, integrations, or availability across plans. Where a feature is described, treat that description as the vendor’s contract; where a feature is not described, do not assume it exists. Scheduling, approval, or export alone does not imply automatic publication; CMS support, draft states, content rules, and exceptions remain documented constraints rather than automatic behavior.

Finally, this guide is documentation-based research, not a live API test. It does not test Frase, does not measure scores, and does not verify undocumented behavior. Any numeric vendor quota or plan detail is omitted here because it is not essential to the whole-word decision.

Final Verdict

The whole-word rule is a small documented mechanic with a large diagnostic effect. When the SEO score lags while EEAT and GEO look acceptable, the first question is not “how do I raise the score” but “is each tracked keyword actually visible to the scorer as a standalone whole word?” The documentation supports that question and the expanded score view supports the audit. It does not support predicting the score change, assigning point values, or treating the score as a ranking guarantee.

Use the keyword whole-word check note as the reusable deliverable. Classify each term, rewrite only where meaning survives, accept gaps deliberately, and re-check the expanded score. That process separates a genuine coverage gap from an embedded term, which is the decision the documentation leaves to the editor. For broader context on the tool, see /frase-review/.

Official Sources

Frase Content Scores Explained: https://docs.frase.io/feature-reference/content-scores-explained.md
Frase Optimize Existing Content: https://docs.frase.io/core-workflows/optimize-existing-content.md