ONE TOPIC. A MORE USEFUL SHORTLIST.
Free Keyword Research Tool
Start with an English topic and explore vocabulary you may want to investigate. The current free mode combines word suggestions with clearly labeled generated ideas. Use the ideas to ask better questions about your audience, content and offer before deciding what deserves a page.
01 / YOUR STARTING POINT
Begin with a subject you can serve well
Enter one seed keyword or topic, such as an actual service, product category or customer problem. A focused phrase usually makes the review easier: “editorial calendar” gives you a clearer starting point than “business.” Choose a subject your team can explain with practical knowledge and accurate examples.
RESEARCH WORKSPACE / ENGLISH IDEAS
Find a useful starting point.
Explore word suggestions and generated phrases, keep their origins, then build a shortlist you can review.
No measured search volume, CPC, keyword difficulty, live SERP or country-specific demand. The seed goes to this website and, when enabled, Datamuse. Generated phrases are not observed searches.
Choose a seed to begin.
CANDIDATES / REVIEW REQUIRED
Keep the origin beside the idea.
How wording, groups and missing metrics are handled
Comparison keys merge normalized spelling and retain original variants and all contributing origins. The report records the exact comparison method. A spelling match does not establish the same search intent.
Variant groups retain English word-stemming relationships. They do not compare search results or decide which phrases belong on one page. Suggested intent is a wording hypothesis; mixed and unknown remain valid choices. Volume, CPC and organic difficulty are unavailable, rather than zero.
Review notes, selection and filters remain in this tab's memory. Downloads contain supplied text and are not automatically anonymized. This is a report workflow, not saved projects or an editable-project import.
A short brief for the next review
Your editorial notes are unverified and remain separate from the source report.
One literal phrase per filter, with NFC normalization, collapsed whitespace and JavaScript default lowercase. Matching is a substring of the displayed phrase; punctuation and accents remain. No stemming or regular expressions. Selection remains when a row is hidden by a filter.
REVIEW HANDOFF
Take the context with the candidates.
JSON retains typed nulls. CSV uses empty metric cells with an explicit unavailable status, and prefixes formula-like text for spreadsheet safety.
No export prepared.
Select and copy the text manually. A prepared download is not confirmation that your browser saved a file.
Seed, market and verification guidance
The visible seed field allows up to 200 UTF-16 code units. Some symbols use more than one unit, so this is not always the same as 200 visible characters. Keep the seed concise and remove account numbers, customer details or confidential campaign information.
Country choices are United States, United Kingdom, Canada, Australia and Worldwide. English is the only language currently offered. The form starts with United States and English selected. Keep your intended audience in your planning notes because the country selector does not turn free suggestions into local demand evidence.
Complete the displayed human verification when you run your own lookup. The page says its code expires after ten minutes and can be used once. Clear resets the seed and returns the observed country and language fields to their defaults. Check those choices before starting another subject.
02 / TRACE THE WORDING
Understand where an idea comes from
The free-mode disclosure identifies Datamuse as the word-suggestion source and says the seed text is sent there. Template ideas are generated locally. Neither source supplies measured search demand in this free mode, and no report upload is needed.
Read the origin label before interpreting any phrase. A word association can suggest vocabulary worth exploring; a generated question can suggest an angle worth discussing with a customer. Neither is proof of a Google autocomplete suggestion, a People Also Ask question or a query from your own Search Console account.
If you want to build phrases from explicit seed variations and modifiers, the Keyword Generator offers a separate drafting workflow. Choose the tool according to the task you are trying to complete, and retain the source of each idea when you move it into a brief.
03 / QUALITY BEFORE QUANTITY
Review relevance before collecting more phrases
Ask whether an idea describes something your audience needs and something your business can provide. A phrase may be linguistically related while pointing to an unrelated product, a different meaning of a word or a task outside your service.
Write one sentence describing the reader and the decision they need to make. For example: “A small editorial team needs a repeatable way to plan and approve articles.” That sentence helps you distinguish a relevant template, workflow guide or comparison from a large list of loosely connected marketing terms.
Keep meaningful modifiers. “Free,” “for schools,” “without software” and “for agencies” can change the required answer. Removing these words to make phrases look similar can erase the reason a reader chose one query over another. Preserve the original wording alongside your review notes.
Reject awkward combinations and offers you cannot support. A generated location phrase does not establish service coverage in that place. A generated price question does not establish your current price. Research helps identify what needs evidence; it does not supply business facts by itself.
04 / KEEP THE DECISION REVIEWABLE
Separate an idea from an intent decision
Intent describes what someone wants to achieve. A reader may need an explanation, a practical resource, a comparison, a supplier or a particular website. The same short phrase can support several interpretations, so an initial label should remain open to review.
Treat any intent you assign during this workflow as your editorial judgment unless you have recorded relevant evidence. A question beginning with “how” often suggests explanation, but its wording alone cannot establish the composition of current search results.
Use the Search Intent Checker when you have supplied search-result evidence to review. Record the source, market, language, device and capture date with that evidence. The free research widget does not fetch those results for you.
When evidence is incomplete, keep the decision pending. You can still record a useful customer question without declaring a dominant intent, choosing a final URL or giving the phrase a numeric opportunity score.
05 / AN AUTHORED EXAMPLE
Worked example: editorial calendar
The following phrases and decisions were written for this document. They are illustrative research candidates, not production output, verified search queries or measured opportunities. All metric values are unavailable in this example.
| Candidate phrase | Authored interpretation | Next review step |
|---|---|---|
| editorial calendar | Broad and potentially mixed | Clarify whether the reader needs an explanation, template or software |
| what is an editorial calendar | Informational hypothesis | Outline a definition and a practical example |
| editorial calendar template | Resource-seeking hypothesis | Check whether a usable template satisfies the task |
| best editorial calendar software | Commercial comparison hypothesis | Define fair comparison criteria and inspect relevant evidence |
| editorial calendar software pricing | Commercial or transactional hypothesis | Verify current supplier pricing and buying requirements |
| editorial calendar for schools | Audience-specific hypothesis | Confirm school workflows and the team’s relevant experience |
Do not sort these candidates by invented volume or mark unavailable demand as zero. The template phrase could be valuable to a relevant audience even before you have a demand estimate. The software comparison may deserve its own experience because the reader needs evaluation criteria and current product evidence.
Your next decision might be to improve an existing guide, develop a resource or hold an idea until you can research it properly. A short list with clear reasons is more useful than a large list whose origin and intended reader have been forgotten.
06 / ONE PAGE OR MORE?
Decide what belongs together
Similar wording can help you find candidates for a shared page, but it cannot finish the page decision. An explanation and a downloadable template may support one experience if the page serves both needs well. A detailed software comparison may require a different format, evidence and maintenance schedule.
Use the Keyword Cluster Generator to organize supplied terms by its supported comparison methods. Review the relationship behind a group before treating every member as a target for the same page. Group membership is a prompt for inspection.
Avoid creating a separate URL for every wording variation. First ask whether an existing page can answer the additional question naturally. Split pages when the reader’s task or the required experience changes enough to justify a useful, distinct destination.
Once you have made that judgment, record page assignments in the Keyword Mapping Tool. Keep the primary task, supporting phrases, intended URL, evidence and owner together. An assignment records a plan; it does not prove that a page exists or performs well.
07 / SUPPORT THE IMPORTANT CALLS
Add independent evidence where it matters
For an important publishing decision, supplement ideas with evidence appropriate to the question. Customer conversations can clarify vocabulary. Your own search performance can reveal queries that already surface your site. A permitted keyword-data source can provide estimates with explicit definitions and context.
Keep estimates separate from observations. Search impressions describe appearances of your site in the reporting source; they are not the total number of searches for a topic. Advertising competition describes a paid-search setting and should not be renamed organic ranking difficulty.
If you later use volume or advertising metrics, record the provider, geography, language, network, period, currency where relevant and retrieval date. Compare compatible records. A broad market estimate and a narrow local observation may both be useful while answering different questions.
Check seasonality and audience fit before committing resources. A temporary event can produce a brief surge in interest. A modest but highly relevant question can support an important customer decision. Neither a large number nor a long phrase automatically identifies the best page to create.
08 / KEEP THE REVIEW GOING
Follow through after publication
Choose a clear objective before publishing: answer a recurring question, help someone complete a task or explain an offer accurately. Record what you changed and when. This gives later review a useful starting point.
For a separate position check, consult the Keyword Rank Checker and its current provider requirements. Its availability and coverage must be assessed separately. A keyword idea does not establish that a rank-data source is connected or that a future position has been measured.
Review outcomes using appropriate first-party evidence and consistent contexts. A ranking change alone cannot establish why it happened. Keep useful content, technical accessibility, competition and business relevance in the discussion rather than attributing every movement to a keyword choice.
09 / BEFORE YOU RELY ON A RESULT
Common questions
Is this a free keyword research tool?
The widget is labeled free mode and provides English word suggestions and generated ideas. Its current disclosure excludes measured volume, CPC, difficulty, live results and local demand. Review that scope before relying on a broader description of keyword research.
Does selecting Canada or Worldwide change the free data?
The free-mode disclosure says country selection does not localize the data. Keep the selected planning market separate from any claim about where an idea is searched. Local demand requires suitable independent evidence.
Are the generated phrases real searches?
Generated phrases are candidates for investigation. Do not label them observed searches, autocomplete results or verified questions unless you have separately collected evidence that supports that description.
Does the tool automatically choose my best keyword?
The inspected free-mode description does not establish a measured best-keyword recommendation. Choose candidates by relevance and evidence, and explain the intended reader, required experience and business reason behind each priority.
Can I interpret a missing metric as zero?
No. Unavailable data means the value has not been supplied or measured in this context. Zero is a specific reported value with its own definition. Keep those states distinct when preparing a brief or comparing sources.
Can keyword research guarantee a ranking?
No. A research candidate does not promise search placement, traffic, leads or sales. Content usefulness, technical conditions, competition and many other factors affect results. Evaluate the work against its purpose and the evidence you can actually collect.
Those order commitments are distinct from ranking or traffic outcomes. Paid-link qualification also needs explicit review: Google’s guidance prefers sponsored for paid placements and accepts nofollow. Resolve the required attributes in the order before approval; keyword research cannot decide them.
A CONCRETE CAMPAIGN BRIEF
Ask about campaign support
Share the target page, audience, market and what you want the work to achieve. Sending a brief requests a review; it does not approve a placement.
