What this tool measures
The lab compares important terms, phrases, and topical signals across both documents and shows where their keyword footprints intersect.
Free Market Brew SEO tool
Free Keyword Separation Lab
Visualize whether two pages or drafts target distinct keyword themes, or whether their language creates avoidable SEO overlap.
How to use the results
Load two URLs or paste two drafts, review shared and unique language, then clarify headings and supporting copy so each page answers a different search need.
Why it matters for SEO
Separating keyword intent helps prevent cannibalization and gives every page a clearer role in your organic search strategy.
Start with a focused SEO question
A useful analysis begins with a specific page and a specific decision. Before opening Free Keyword Separation Lab, write down the search intent the page should satisfy, the audience that needs the answer, and the conversion or next step that matters. This prevents a common mistake: changing copy simply because a chart looks weak. The visualization is evidence, not a writing instruction by itself. The lab compares important terms, phrases, and topical signals across both documents and shows where their keyword footprints intersect. Read the result in that context, then decide whether the evidence supports the page's intended job. A product page, educational article, location page, comparison page, and support document can legitimately show different patterns even when they discuss the same broad topic.
Use the exact canonical URL whenever possible. Redirected URLs, tracking parameters, alternate hostnames, and outdated staging addresses can point at a different document than the one search engines evaluate. Confirm that the visible title, headings, main copy, and page purpose match what you expect before interpreting a score. If the tool accepts a search phrase, choose the phrase whose results you are actually trying to improve. A broad category phrase and a detailed question can produce very different evidence because they represent different needs.
Read the visualization from evidence to action
Begin with the labeled fields, highlighted passages, or scored components rather than the overall result. Those details show which parts of the document contributed to the measurement. Look for a consistent explanation across the title, description, headings, introductory copy, supporting sections, and conclusion. Strong pages usually make their subject clear early, develop it with useful detail, and keep supporting material connected to the main purpose. A weak component is a place to investigate, not an automatic command to add a phrase or repeat a term.
Load two URLs or paste two drafts, review shared and unique language, then clarify headings and supporting copy so each page answers a different search need. Translate that workflow into one small hypothesis. For example, you might clarify an ambiguous heading, add a missing explanation, move an important answer closer to the section that introduces it, remove unrelated boilerplate, or make the relationship between two concepts explicit. State what you expect the visualization to show after the edit. A measurable hypothesis is more useful than a vague goal such as “optimize the page,” because it tells you what changed and why.
Use comparisons as a diagnostic, not a template
When a comparison URL is available, use it to understand differences in structure and evidence. The other page is not automatically correct, and it should never be copied. It may rank for reasons outside the scope of this visualization, such as authority, links, freshness, brand demand, location, or a better match to the current results page. Compare the labeled components and ask what useful information the other document communicates more clearly. Then express that information in a way that is accurate, original, and appropriate for your own audience.
A good comparison keeps the search phrase and analysis method constant. If you change the target, comparison, and phrase at the same time, you will not know what caused the difference. Review one variable at a time. Pay attention to both strengths and weaknesses: the target may already explain some concepts better than the outperformer. Preserve those strengths while fixing the specific gap shown by the evidence. The goal is a better page, not a page that merely resembles another URL.
Measure an edit before publishing
Use the editable preview when it is available. Change one meaningful passage, heading, or field, and then measure the revision. The refreshed visualization should make the effect visible through updated highlights, component values, or scores. If nothing changes, confirm that you edited a field used by this particular analysis and that the edit contains the idea you intended to test. Some measurements evaluate literal words, some evaluate relationships or distance, and others evaluate semantic meaning. An edit that affects one method may correctly have little effect on another.
Keep a short record of the original evidence, the edit, the new evidence, and the reason for publishing or rejecting it. This turns a one-time visualization into a repeatable editorial process. Rejected experiments are useful because they show which changes did not improve the page's measured relationship to the search intent. Accepted experiments should still be reviewed for accuracy, readability, brand voice, accessibility, legal requirements, and conversion impact before they reach production.
Avoid score chasing and keyword stuffing
A higher value is not always a better reader experience, and no single factor determines organic performance. Do not force the search phrase into every heading, add unsupported claims, or inflate a page with repetitive paragraphs. Search systems evaluate many signals together, while people quickly notice awkward language and unnecessary repetition. Use Free Keyword Separation Lab to locate a communication problem, then solve that problem with concise, helpful content. The strongest revision often adds specificity, evidence, examples, definitions, or clearer organization rather than more occurrences of the same words.
Separating keyword intent helps prevent cannibalization and gives every page a clearer role in your organic search strategy. That benefit is most reliable when the page remains useful on its own. Review the complete document after making a localized change. Confirm that transitions still make sense, important qualifications remain intact, links point to helpful supporting resources, and the primary answer is easy to find. If an edit improves a component score but makes the page harder to understand, revise it again or discard it.
Build a repeatable optimization checklist
First, verify the target URL, search intent, and page purpose. Second, inspect every labeled component and identify the weakest evidence that is relevant to that purpose. Third, compare the target with an eligible outperformer when comparison data is available. Fourth, write one concrete change and predict its effect. Fifth, measure the edited preview. Sixth, review the result for human clarity and factual accuracy. Finally, publish only the changes that improve both the evidence and the page. Revisit the analysis after the page is crawled again so the stored document reflects what users can actually see.
Treat this checklist as quality control rather than a rigid formula. Different queries require different kinds of answers, and different page types have different constraints. The value of the tool is that it exposes how a defined scoring process reads the supplied content. Your expertise supplies the context the score cannot know: what is true, what the audience needs, what the business can promise, and what belongs on this particular page. Combining transparent measurement with editorial judgment creates changes that are easier to explain, test, and maintain.