🔍 Free Keyword Research Tool
Discover keyword ideas, search intent, smart clusters and autocomplete signals — completely free
✨ Alphabet Soup Method📚 Complete User Guide
Learn how to discover, understand, filter, cluster and export autocomplete-based keyword ideas.
🎯 What Is This Free Keyword Research Tool?
This free keyword research tool helps you discover keyword ideas from autocomplete suggestions across Google, YouTube, Bing and Amazon. Enter a seed keyword, choose a platform and language, and the tool generates a structured list of related suggestions.
The tool combines autocomplete discovery with local analysis. It can classify search intent, identify keyword types and attributes, calculate Seed Relevance, measure an Autocomplete Signal, and organize results into smart clusters — without requiring a paid keyword-data API.
Autocomplete is a discovery signal. It is not a replacement for search-volume, keyword-difficulty, traffic, ranking or conversion data from dedicated SEO platforms.
4 Platforms
Google, YouTube, Bing and Amazon autocomplete discovery
10 Languages
Research keywords across supported languages and markets
Smart Analysis
Intent, types, attributes, relevance, signals and clustering
Free Analysis
Local analysis with no paid keyword-data API required
🚀 How to Use the Tool: Step-by-Step
Enter Your Seed Keyword
Enter the main topic you want to research. The seed keyword is used as the starting point for autocomplete queries and for the local relevance analysis.
- Use a clear topic or phrase relevant to your niche.
- Specific phrases such as “email marketing” can produce more focused results than very broad terms.
- You can also test shorter seeds when exploring a new topic.
- Run several related seeds when you want broader topic coverage.
“digital marketing” · “fitness tips” · “python tutorial” · “healthy recipes” · “EPFO wage ceiling”
Select Platform and Language
Choose the autocomplete source that matches your research objective, then select one of the supported languages.
- Google: General web-search autocomplete discovery.
- YouTube: Video-search autocomplete discovery.
- Bing: Bing autocomplete discovery.
- Amazon: Product and shopping-oriented autocomplete discovery.
Arabic, English, French, German, Hindi, Italian, Japanese, Portuguese, Russian and Spanish.
Run the same seed on different platforms when you want to compare how the topic appears in web, video or product-oriented autocomplete.
Choose Generation Options
Use the checkboxes below the search field to control which query variations are generated.
- A-Z Variations: Adds alphabet-based variations to systematically explore autocomplete suggestions.
- 0-9 Variations: Adds numeric variations that can uncover year, list, version and other number-related phrases.
- Question Keywords: Adds question-oriented queries such as how, what, why, when, where, who, which, can, will and should.
- Auto-filter Low Quality: Applies the tool's local filtering rules after analysis to reduce weaker or less specific results.
The tool starts with the seed and adds the selected variation queries. With A-Z, 0-9 and Question Keywords all enabled, the current implementation can process up to 57 query variations before duplicate suggestions are merged.
Generate and Wait for Analysis
Click Generate Keywords. The progress area shows the number of queries processed and keywords found while the selected autocomplete source is queried.
- Autocomplete suggestions are collected from the selected source.
- Duplicate keywords are merged.
- The number of autocomplete query variations in which each keyword appeared is tracked.
- Local analysis assigns intent, type, attributes, relevance, signal and cluster information.
The final number of results is not fixed. It depends on the seed keyword, platform, language, available autocomplete suggestions, duplicate overlap and the selected filtering options.
Understand the Result Cards
Each result card separates the original keyword from the tool's generated analysis so you can see both the discovery phrase and its classification.
- Keyword: The actual autocomplete phrase discovered by the tool.
- Search Intent: Informational, Commercial, Transactional, Navigational or News / Current.
- Keyword Type: Core, Mid-tail, Long-tail, Question, PDF / Document, Comparison, Year-specific or Location / Language.
- Attributes: Additional characteristics detected in the phrase, such as Question, PDF / Document, Comparison, Year, Location / Language, Current / News or Notification / Circular.
- Seed Relevance: A 0–100 score representing topical closeness to the seed.
- Autocomplete Signal: Strong, Good, Moderate or Low based on repeated appearance across the tool's generated autocomplete queries.
- Appeared ×N: How many generated autocomplete queries returned that keyword.
- Cluster: A human-readable topic grouping generated locally from the keyword and seed.
Filter, Sort, Cluster and Generate Content Ideas
After the results are ready, use the analysis controls to focus on the part of the keyword set you need.
- Intent Filter: Show only a selected search-intent category.
- Demand Signal Filter: Filter by Strong, Good, Moderate or Low autocomplete signal.
- Filter Keywords: Search within the generated result list by text.
- Sort by Score: Order results using Seed Relevance, then supporting signals.
- Sort by Frequency: Order by autocomplete appearances.
- Sort A-Z: Alphabetical ordering.
- Sort by Length: Order by keyword length.
- Show Clusters: View the detected parent clusters and filter results by cluster.
- Generate Content Ideas: Create local blog, YouTube, FAQ, keyword-opportunity and commercial-topic ideas from the seed and analyzed results.
- Copy All: Copy the keyword list to the clipboard.
- Export CSV: Download the analyzed dataset for spreadsheet work.
🎯 Understanding Search Intent, Relevance and Signals
Search Intent
The tool classifies keywords into Informational, Commercial, Transactional, Navigational and News / Current based on the wording and detected patterns.
Seed Relevance
This 0–100 score estimates how closely the discovered keyword matches the seed topic. It is not search volume, traffic, difficulty or ranking probability.
Autocomplete Signal
This is an internal discovery signal based on repeated autocomplete appearances during the current research run. It is not monthly search volume.
Understanding Seed Relevance
Exact or Very Close Match
Keywords that closely match the seed receive stronger relevance because they are directly aligned with the topic entered.
Seed Word Coverage
Keywords containing more of the important words from the seed are generally treated as more closely related.
Structural Signals
Question, year, document/notification and repeated-autocomplete characteristics can contribute small adjustments to the local score.
What the Score Does Not Mean
A higher Seed Relevance score does not mean higher search volume, lower competition, more traffic, better rankings or higher conversion potential.
Understanding Autocomplete Signal
Strong Signal — ×5 or More
The keyword appeared in at least five of the autocomplete query variations processed in this run.
Good Signal — ×3 to ×4
The keyword appeared in three or four generated autocomplete queries.
Moderate Signal — ×2
The keyword appeared in two generated autocomplete queries.
Low Signal — ×1
The keyword appeared once in the generated autocomplete queries.
If a keyword shows ×6, it means the tool found that keyword in six of its generated autocomplete queries. It does not mean six people searched for it, six monthly searches, or a search-volume estimate of six.
⚙️ Keyword Types, Attributes and Smart Clustering
Keyword Types
The tool assigns a primary type to help organize the result set.
- Core
- Mid-tail
- Long-tail
- Question
- PDF / Document
- Comparison
- Year-specific
- Location / Language
Keyword Attributes
Attributes add useful secondary context to a keyword.
- Question
- PDF / Document
- Comparison
- Year-specific
- Location / Language
- Current / News
- Notification / Circular
Smart Clustering
Keywords are grouped into readable topic clusters based on the seed and important terms in each phrase.
- Parent cluster identifies the main topic angle.
- Subcluster adds a more specific qualifier.
- Amount transitions can be grouped separately.
- Notifications, calculations, proposals, history and impacts can receive distinct labels.
Alphabet Soup Method
A-Z Variations systematically expands the seed with alphabet-based queries to expose autocomplete suggestions you may not think of manually.
- Useful for broad topic discovery
- Can reveal long-tail phrases
- Works with supported languages
- Results depend on available autocomplete suggestions
Numeric Variations
0-9 Variations adds numeric query patterns that may reveal years, numbered content, versions and other number-related searches.
- Year-specific phrases
- Numbered guides
- Version-related queries
- Other numeric modifiers
Question Keywords
Question expansion helps discover question-shaped queries that can be useful for articles, FAQs, tutorials and videos.
- How, what, why, when and where
- Who, which, can, will and should
- Useful for FAQ research
- Useful for educational content ideas
Auto-Filter
The optional filter reduces weaker results using the current local rules rather than claiming that a keyword is objectively good or bad.
- Keeps Strong and Good signals
- Keeps repeated suggestions
- Keeps sufficiently relevant results
- Retains multi-word phrases
Multi-Language Research
Use the language selector to research autocomplete suggestions in Arabic, English, French, German, Hindi, Italian, Japanese, Portuguese, Russian or Spanish.
- Useful for international topics
- Compare wording across languages
- Research local-market phrasing
- Combine with platform comparison
💼 Common Use Cases
Content Creators & Bloggers
Discover article topics, question-based ideas, long-tail variations and related angles around a seed topic.
YouTube Creators
Use YouTube autocomplete to discover video-oriented wording, tutorials, reviews and question-based topics.
SEO Research
Build a starting keyword set, group related topics, identify intent and export the results for further SEO validation.
Product & E-commerce Research
Amazon autocomplete can help uncover product-oriented phrases, features, comparisons and buying-related wording.
Content Planning
Use clusters, intents, questions and generated content ideas to organize a content calendar around a topic.
Keyword List Building
Export the analyzed results to Excel or another spreadsheet for further filtering, tagging and planning.
💡 Practical Tips for Better Research
Test Several Seeds
Use closely related seed phrases instead of relying on one seed to represent an entire topic.
Compare Platforms
Google, YouTube, Bing and Amazon can surface different autocomplete wording because they serve different search contexts.
Use Question Expansion
Enable Question Keywords when you need FAQ, tutorial, educational or problem-solving ideas.
Review Signal in Context
A Strong Signal means repeated appearance during this run. It does not establish search volume or popularity by itself.
Use Clusters to Organize
Smart clusters can help turn a long keyword list into manageable topic groups and subtopics.
Use Seed Relevance Correctly
Use the score to understand topical closeness to your seed, not as a substitute for SEO metrics.
Export for Deeper Analysis
Use CSV when you need spreadsheet formulas, notes, external search-volume data or your own prioritization.
Validate Before Publishing
Check search intent, competition, business relevance and current information before turning a keyword into content.
✅ Best Practices
✓ DO
- Use a seed that clearly represents your topic.
- Compare multiple platforms when relevant.
- Use A-Z, numeric and question expansion according to your research goal.
- Review intent, type, attributes and clusters together.
- Use Autocomplete Signal as a discovery signal, not search volume.
- Use Seed Relevance as a topical-match indicator.
- Export important research sessions for further analysis.
- Validate volume, competition and business relevance with additional data before making SEO decisions.
✗ DON'T
- Assume ×5 means five searches or five monthly searches.
- Treat Seed Relevance as keyword difficulty or traffic.
- Assume every autocomplete suggestion is equally important.
- Assume a fixed number of keywords will be returned for every seed.
- Use the tool as a replacement for dedicated search-volume data.
- Assume a long-tail keyword automatically has low competition.
- Ignore the platform context when interpreting a suggestion.
- Publish or target a keyword without reviewing its actual intent and context.
❓ Frequently Asked Questions
What data does this tool use?
The tool collects autocomplete suggestions from the selected platform and language. The current implementation supports Google, YouTube, Bing and Amazon. Additional analysis such as intent, relevance, type, attributes and clustering is calculated locally in the browser.
Does the tool provide search volume?
No. The tool is designed for keyword discovery. The Autocomplete Signal and Appeared ×N values describe how often a keyword appeared across the autocomplete query variations generated during your current run. They are not monthly search-volume estimates.
What does “Appeared ×6” mean?
It means the same keyword was returned by six of the autocomplete query variations processed for that research run. It does not mean six searches, six users or six monthly searches.
What is Seed Relevance?
Seed Relevance is a 0–100 local score that estimates how closely a discovered keyword matches the seed topic. It considers the relationship between the keyword and seed, including seed-word coverage and selected structural signals. It is not search volume, keyword difficulty, traffic or ranking probability.
What are Strong, Good, Moderate and Low signals?
- Strong: appeared 5 or more times.
- Good: appeared 3–4 times.
- Moderate: appeared 2 times.
- Low: appeared once.
These levels describe repeated autocomplete appearances during the current run.
What is the Alphabet Soup Method?
It is a systematic autocomplete-discovery approach that expands a seed with alphabet variations. In this tool, A-Z Variations adds alphabet-based query variations so the selected platform can return additional suggestions. It is a discovery technique, not a search-volume calculation.
How do 0-9 Variations work?
The numeric option adds 0–9 query variations. These can uncover year-related phrases, numbered content, versions and other numeric modifiers that occur in autocomplete.
What are Question Keywords?
The question option expands the research with terms such as how, what, why, when, where, who, which, can, will and should. These queries are useful when you want question-shaped content ideas, FAQs or educational topics.
How does Auto-filter work?
Auto-filter applies the tool's current local rules after analysis. It keeps Strong and Good autocomplete signals, repeated suggestions, sufficiently relevant results and multi-word phrases. Because filtering is rule-based, you can turn it off when you want to inspect a broader set of raw discoveries.
What are Smart Clusters?
Smart Clusters group keywords into readable topic angles based on the seed and important terms in each keyword. For example, a phrase about a wage-ceiling proposal can be grouped under Proposal / Consideration, while a notification or calculation phrase can receive a different cluster. Clustering is an organizational aid, not a claim about search volume or ranking.
Why does the tool show both a keyword and an analysis label?
The bold keyword is the original autocomplete phrase discovered by the tool. The smaller analysis line is a generated interpretation that makes the topic angle easier to scan. Keeping them separate prevents the generated label from being mistaken for the actual search phrase.
Why can two platforms return different keywords?
Each platform has its own autocomplete system and search context. Google, YouTube, Bing and Amazon can therefore return different suggestions for the same seed. Differences are useful for discovering platform-specific wording, but they should not be interpreted as directly comparable search-volume measurements.
How many keywords will I get?
There is no fixed result count. The number depends on the seed, platform, language, selected query options, autocomplete responses, duplicate overlap and Auto-filter. With all generation options enabled, the current implementation can process up to 57 query variations before duplicate suggestions are merged.
Can I use the results for SEO?
Yes, as a keyword-discovery starting point. You can use the results for content ideas, FAQs, topic grouping, video planning and further SEO research. Before prioritizing a keyword, validate search volume, competition, current relevance, search intent and business value with appropriate additional sources.
Can I export the analyzed results?
Yes. Copy All copies the keyword list to your clipboard, while Export CSV downloads the analyzed dataset. The CSV includes Keyword, Intent, Keyword Type, Attributes, Parent Cluster, Subcluster, Autocomplete Signal, Autocomplete Appearances and Seed Relevance.
What are the Content Ideas?
The Content Ideas feature uses the current seed and analyzed results to create local suggestions for Blog / Article Ideas, YouTube Ideas, FAQ Ideas, Keyword Opportunities and Commercial Topics. These are generated ideas, so review them for accuracy and relevance before publishing.
Do I need a paid keyword-data API?
No paid keyword-data API is required for the local analysis features. The tool uses the selected autocomplete sources for discovery and performs intent, relevance, classification and clustering locally.