AI Keyword Cluster Generator

Group up to 100 keywords by topic and likely intent. Review suggested clusters, unassigned terms and the original keyword list, then export CSV.

Options

Screpy’s AI keyword cluster generator organizes an existing list of keywords into suggested groups by meaning and likely intent. It keeps the original terms visible, identifies a representative keyword for each group and leaves unsuitable terms unassigned. Use the groups to review your content plan before deciding which queries belong on the same page. They are semantic suggestions, not live Google research.

Prepare your keyword list

Paste 2–100 keywords, one per line, with 2–100 characters in each term. Start with an existing research list or queries relevant to your site. The tool groups what you supply; it does not discover additional keywords from a seed topic.

Blank lines are ignored. Duplicate detection normalizes case, spacing and compatible Unicode forms, then keeps the first version of a repeated term. You need at least two unique terms after this process. The 100-line limit applies to the submitted nonblank list, so duplicates do not let you submit a larger batch. An oversized list is rejected, not silently shortened.

In Options, add up to 500 characters of website or market context when a term has several meanings. Choose one of 11 languages for group labels and explanations. The original keywords are not translated or rewritten. Keep a copy of your source list if you want to compare different grouping runs.

Understand clusters, primary terms and unassigned keywords

Each group includes a label, a suggested primary keyword, a likely intent and a short explanation. The primary keyword represents the group; it is not a verified search-volume leader or a recommendation to prioritize that page first.

Groups appear in descending size, with ties resolved by the original keyword order. Terms inside a group follow their input order. Neither position is an opportunity score. A group may contain just one term when combining it with others would be unsuitable.

Every unique keyword must appear exactly once, either in a group or in the unassigned list. Unassigned terms remain visible with an ambiguous or unrelated explanation. Review these rather than deleting them automatically: an unclear term may need better context, while an unrelated one may belong in another project.

Review a grouping example before mapping pages

Imagine a folding-bike site supplies six terms: “folding bike”, “folding bicycle”, “best folding bike”, “folding bike price”, “folding bike hinge repair” and “java”. This fictional editorial grouping shows how to assess a result; it is not an observed tool output or SERP comparison.

Proposed group Supplied terms Editorial question
General folding bikes folding bike; folding bicycle Can the same page satisfy these wording variants?
Choosing and pricing best folding bike; folding bike price Do readers want a comparison, a product listing or a budget explanation?
Hinge repair folding bike hinge repair Does this need a separate troubleshooting answer?
Unassigned java What does the term mean, and is it relevant to this site?

The shared phrase “folding bike” does not settle the page decision. Someone researching a purchase and someone repairing a hinge need different answers. Even the choosing-and-pricing group may need to be split after reviewing the actual searches and the site’s existing content.

For an ambiguous term, supply useful context or remove it from this particular project only after checking its origin. A group is a starting hypothesis, not an instruction to create, merge or redirect pages.

Export a plan you can verify

Before assigning a group to a URL, check the reader’s task, current search results for the intended market and existing pages that already answer the question. This tool supplies no search volume, keyword difficulty, SERP overlap or cannibalization analysis. If those signals are needed for a decision, collect them separately; do not infer them from group size or wording similarity.

Use the AI Search Intent Analyzer to explore a single query’s likely purpose, while remembering that its classification also needs verification. Once you accept a topic and page scope, the AI Content Brief Generator can help turn them into writer instructions.

Copy the groups as Markdown or download CSV containing keywords, clusters, primary status and suggested intent. Unassigned terms are included in the export. Keep your own notes about the intended page, evidence and next action alongside the exported data.

Google’s SEO Starter Guide explains that a page can relate to query variations without using every exact phrase. Write a useful answer naturally instead of forcing every clustered term into the text.

A valid email is required for each request. Form values and email are recorded with tool usage; email is excluded from the AI prompt. Inputs stay in this browser and are restored on return without automatically running the tool. Avoid confidential information.

Frequently asked questions

Group up to 100 keywords by topic and likely intent. Review suggested clusters, unassigned terms and the original keyword list, then export CSV.

Is this tool free?

Yes. A valid email is required to run it; request limits apply.

Why are some terms unassigned?

Their meaning may be ambiguous or unrelated. They remain visible and are included in CSV export.

What does primary keyword mean?

It is a suggested representative of the group, not the term with verified highest volume.

Is my input saved?

Form values stay in this browser and are recorded with your email in usage logs. Text is processed by AI; avoid confidential information.

Does clustering use live SERPs?

No. It uses semantic AI inference from your list and optional context.

Can keywords disappear or be added?

The API rejects missing, repeated or invented assignments. Original text is restored from your input.