Keyword Grouping Tool

A keyword grouping tool turns a long, messy list of search terms into tidy clusters you can plan pages around. Paste your keywords, one per line, or upload a CSV such as a Google Search Console queries export or a Keyword Planner download. The tool lowercases and cleans every keyword, sets aside filler words, folds plurals together and groups keywords that share the same important terms. Each group gets a suggested name, a keyword count and summed volume, impressions and clicks, so you can see at a glance which topics matter most. A separate view sorts keywords by intent, such as informational, commercial, local and branded. Rename and merge groups, search, then export a CSV. Everything runs in your browser, and lists of around 10,000 keywords usually group in a second or two.

How to use the Keyword grouping

1. 1 Add your keywordsPaste a plain list, one keyword per line, or switch to CSV with metrics and upload an export. The tool finds the header row and guesses the keyword, volume, impressions, clicks and position columns. Check the guesses and change any that are wrong.
2. 2 Tell it about brands and placesEnter your own brand names and your competitors' brands, plus any city or region names you target. Keywords that contain them are marked as branded or local, and those words no longer split groups.
3. 3 Choose a strictnessBalanced suits most lists. Move to Broad for fewer, larger topic groups, Strict when groups mix things that should stay apart, or Close variants to find near-duplicates such as plurals and reordered words.
4. 4 Review the groupsOpen a group to see its keywords, each with its intent and metrics. Hover a keyword to see which terms placed it there. Rename groups to match the page you will write, and merge groups that belong on one page.
5. 5 Export and planDownload the CSV with one row per keyword, the group summary with one row per group, or copy everything into a spreadsheet. Then map each group to a single page on your site.

Why group keywords at all

Keyword research tends to produce lists that are far too long to act on one line at a time. A Search Console export for a modest site can hold thousands of queries, and a Keyword Planner download for a single seed term can easily run to several hundred. Most of those rows are not separate ideas. "Running shoes for women", "women's running shoes" and "best running shoes for women" all describe the same need, and one well-built page can serve all three.

Grouping is the step that turns a list into a plan. Once similar keywords sit together, you can see which topics carry the most demand, which ones you already rank for, and where the gaps are. It also keeps you from writing several thin pages that chase slightly different wordings of the same search, which usually means none of them performs as well as one strong page would.

Groups are useful beyond new content, too. They help you decide which existing page should own a topic, which internal links to add, and which pages to combine when two of them are competing for the same searches.

How this tool groups keywords

This is lexical clustering: keywords are grouped by the words they contain. The process is deterministic, so the same list and the same settings always produce the same groups, and every decision can be explained. Here is what happens to each keyword.

• Normalize: the keyword is lowercased, apostrophes are removed, other punctuation becomes a space and repeated spaces collapse. "Women's Running-Shoes" becomes "womens running shoes". Keywords that are identical after this step are treated as duplicates and merged.
• Detect modifiers: the tool looks for intent words and phrases, such as how, what, best, cheap, price, review, vs, near me and login, along with the brand names and places you entered. These set the keyword's intent. With "Group by topic, not modifier" switched on, they are then set aside, so "best running shoes" and "cheap running shoes" can share a group.
• Remove filler words: common English words such as the, for, to and of are dropped, because they say little about the topic.
• Fold plurals: simple plural rules map shoes to shoe, boxes to box and companies to company. Endings such as -ing and -ed are left alone, so running and run stay different words. This keeps the results predictable.
• Find shared terms: every combination of one, two or three remaining terms is counted across the whole list. Each keyword joins the most specific combination that enough other keywords also contain, choosing the most common one when there is a tie. Groups that end up smaller than your minimum size are dissolved, and their keywords try the next best combination or land in Ungrouped.
• Name the group: the suggested name uses the words the group's top keyword shares with every member, in that keyword's own word order. The top keyword is the one with the most volume, impressions or clicks, or the shortest keyword when there are no metrics.

Choosing strictness and group size

Strictness controls how much keywords must have in common. Broad looks at one or two shared terms, which gives a short list of large topic groups and is handy for a first look at a big export or for planning site sections. Balanced prefers three shared terms and falls back to two or one, which tends to produce groups the size of a single page. Strict requires at least two shared terms, so single-word overlaps never join keywords together. Close variants only groups keywords whose remaining terms are exactly the same, which surfaces plurals, reordered phrases and modifier variations of one search.

The minimum group size sets how many keywords a group needs. At two, any pair of related keywords can form a group. Raising it to five or more hides small clusters and pushes their keywords into broader groups or into Ungrouped, which helps when you only want to see topics with real depth. If one word appears in almost every keyword, such as your product category or your own site name, add it to "Extra words to ignore" so the rest of each keyword decides the grouping.

Reading the intent view

Switch to the intent view to see every keyword sorted into a single intent bucket. A keyword gets the first match in this order: Branded if it contains one of your brand terms, Local if it contains near me, nearby, nearest or one of your places, Commercial for words such as buy, price, cost, cheap, best, vs, review, compare, deal or for sale, Informational for question words and terms such as guide, tutorial, tips and meaning, and Navigational for login, sign in and similar. Anything else is listed under No modifier.

The matched word is shown next to each keyword, so you can always see why it landed where it did. The rules are simple on purpose, which means they will sometimes misjudge a keyword. "Best" usually signals someone comparing products, but not always, and many keywords with no modifier at all are commercial in practice. Use the buckets as a starting point for the kind of page each topic needs, and check the actual search results for your most important groups before you commit to a format.

What lexical grouping cannot tell you

Some paid platforms group keywords by comparing the search results themselves: if two keywords return many of the same ranking pages, the search engine probably treats them as one topic. That method needs live results data for every keyword, which costs money at scale. This tool does not use it, and it does not look up search results at all.

Grouping by words is fast, free and transparent, but it has blind spots. Synonyms with no shared words, such as "sneakers" and "trainers", will not be grouped unless you merge them yourself. Keywords that share words but mean different things can end up together. And two keywords that look nearly identical may still deserve separate pages if searchers want different things, such as a product list for one and a how-to guide for the other.

The fix is a quick human review. Search a few keywords from each large group and see whether the same kind of page ranks for them. Merge groups that clearly belong together, split out anything that does not fit, and rename groups so each one describes a page you could actually write.

Turning groups into a content plan

The goal is one page per group. Pick a primary keyword for each group, usually the one with the most volume or impressions, and use the rest of the group as the subtopics, headings and questions that page should cover. A group of question keywords often maps neatly onto an FAQ section or a series of headings in a guide.

Then check each group against your existing pages. If a page already covers the topic, improve it rather than writing a new one. If two of your pages already target keywords from the same group, that is a sign of keyword cannibalization: they compete with each other and can split clicks and links. Combine them into one stronger page, redirect the weaker URL, and point your internal links at the page that remains.

Search Console data makes this especially useful. Upload a queries export with clicks, impressions and position, and the group totals show which topics already earn attention and which ones rank on page two. The Keyword Opportunities report in your Make It Flow account works from the same Search Console data, highlighting queries in striking distance, rankings with a low click rate and question searches, and grouping those queries is a natural next step for deciding which page should win each topic.

Metrics, file formats and privacy

When you upload a CSV, the tool sums volume, impressions and clicks for each group. Position is averaged rather than summed, weighted by impressions when they are available, because adding positions together means nothing. When the same keyword appears twice, clicks and impressions are added together, volume keeps the higher figure, and position is averaged. Range values such as "1K – 10K" are read as their lower bound, so treat range-based volumes as rough.

The parser handles comma, semicolon and tab separated files, quoted values, thousands separators and decimal commas, and the UTF-16 encoding that some Keyword Planner exports use. It skips title rows above the header when it can recognise a keyword or query column. Up to 20,000 rows are grouped in one run; split larger lists into batches.

Your keywords never leave your device. Parsing, grouping and exporting all happen in your browser, so nothing is uploaded or stored, and refreshing the page clears everything. Merges and renames are also lost if you group again, so download your CSV before changing settings.

Examples you can copy

Input:
running shoes for women
womens running shoes
best running shoes for women
trail running shoes
best trail running shoes
how often to replace running shoes
when to replace running shoes

Groups:
womens running shoes    3 keywords
replace running shoes   2 keywords
trail running shoes     2 keywords

Top queries,Clicks,Impressions,CTR,Position
running shoes,120,9800,1.2%,8.4
best running shoes,95,7600,1.3%,6.1

Keyword column:  Top queries
Clicks:          Clicks
Impressions:     Impressions
Position:        Position (averaged per group, weighted by impressions)

keyword,group,intent,intent_modifier,impressions,clicks,position
running shoes for women,running shoes women,No modifier,,4100,60,7.9
best running shoes for women,running shoes women,Commercial,best,2900,38,5.5
how to clean running shoes,running shoes,Informational,how,2200,26,3.1
running shoes near me,running shoes,Local,near me,900,6,16.4

Common mistakes to avoid

Frequently asked questions