Tools · SEO Bundle
Keyword grouping tool for a clean core
Collect keywords from every tool into one table, clean them with stop words and word forms, and sort them into groups — a ready core for your brief and text.
A keyword grouping tool turns a raw query list into the structure of your future pages. In SEO Bundle this is the collection table: a single workspace where keywords from every tool pile up and where the core is cleaned before you write a single line. Instead of a dozen exports scattered across tabs, you get one table with groups, inline editing and filters.
The collection pulls data from Serpstat, Google autocomplete and competitor crawling into one de-duplicated list. From there you sort queries into clusters, cut the noise with stop words and fold word forms together — and end up with a clean semantic core that feeds your brief and text.
What the collection table is
The collection table is the semantics accumulator inside the program. Any tool — Serpstat collection, Google suggestions, LSI extraction, competitor overlap — sends the queries it finds into one shared list. Each keyword keeps its volume, cost per click, competition and source, so grouping decisions rest on numbers rather than guesswork.
The list de-duplicates automatically: the same query coming from different sources is never doubled. Columns sort and filter, so thousands of rows turn into a readable picture within a couple of clicks.
Keyword grouping by clusters
On the left is the groups panel. A group here equals a future page or section: you create a cluster and drag the selected rows straight into it with the mouse. You can select a block of queries by dragging, a range with Shift or single rows with Ctrl, then send everything into the right group in one move or through the context menu.
Each group shows its query count and total volume at once, so it is clear which cluster deserves a separate page and which is better merged. That way keyword grouping stops being manual drudgery in an Excel sheet and takes minutes instead of hours.
Cleaning the core with stop words and word forms
Before grouping, the core is cleared of junk. A click on an unwanted word in any row turns it into a stop word, and every query with that stem is reversibly hidden — a counter shows how many rows are tucked away, and one button brings them back or deletes them for good. Stop words match the word stem, not the exact spelling, so they cut all forms at once.
A separate Word Forms window splits the core by stems: "buy" and "buying", "price", "prices" and "pricing" gather into groups. You uncheck an unneeded form or a whole group, while union and subtraction modes either highlight the queries you want or strip out everything else. It is the quick way to remove duplicates by word form that usually bloat the core.
An interactive table in the spirit of Key Collector
The collection is modelled on Key Collector but without the bulk. Any cell is edited in place: a double click or Enter opens the field and changes save instantly. Rows are removed by the cross on hover, the Del key or the context menu — and you can delete either the single clicked row or every selected row at once.
- Row selection one by one, by range with Shift and by mouse-drag block;
- Block copy of the chosen cells to the clipboard as a table for Excel;
- Sorting on a header click and multi-sort across several columns;
- A mini search under each column, including regular expressions;
- Context menu: edit, delete, to group, to stop word, copy.
A click on a shortened site address copies the full URL, and a click on a word in the LSI results shows which pages it appeared on. Small things that save time on every core.
Where the collected data goes
The collection is not a dead end but the entrance to the rest of the pipeline. A cleaned, grouped core feeds straight into the copywriter brief: the selected keywords land in the technical task together with volume and structure recommendations. The same data is available in LSI keyword search and in competitor analysis.
Page crawling is shared across all tools: competitors loaded once are reused between LSI, headings and overlap, so the same pages are not downloaded again for every tab. If you would rather have the work done for you, we will build the core as a service — see keyword research.
When the collection table helps most
The bigger the niche, the clearer the payoff. The semantics of an online store is thousands of queries from Serpstat and suggestions that, without cleaning, turn into a mess: form duplicates, other brands, informational tails mixed with commercial ones. The collection table brings them into one list, while stop words and word forms leave only what is relevant to the topic and the page type in a few passes.
For a content project or blog, grouping matters more: similar queries gather into clusters and you can see at once where one article is enough and where topics should be split across separate pieces. Agencies and freelancers benefit from preparing the core in batches — a single cleaning and grouping routine keeps quality consistent across projects rather than depending on end-of-day fatigue.
A further bonus is repeatability: a saved core can be reopened and cleaned up later without collecting the data again, and the shared crawl saves time on every following tool.
Frequently asked questions
How is grouping in the collection table better than doing it in Excel?
In the collection table the queries are already gathered from different sources, de-duplicated and carry their volume. Groups are built by dragging, and each cluster shows its size and reach. In Excel you would have to do this with formulas and manual copying, losing the volume and source data along the way.
Do stop words delete queries permanently?
No. By default the matched rows are hidden and can be restored at any time — a counter shows how many are hidden. Permanent deletion is a separate button you press deliberately, once you are sure of the result.
How does the Word Forms window work?
It splits the core by word stems and gathers the forms into groups. You uncheck an unneeded form or a whole group, and the table highlights or removes the matching queries. It is handy for clearing duplicates by case and number in a minute.
Where does the data in the collection table come from?
From the program's tools: Serpstat collection, Google autocomplete, LSI search and competitor overlap. Any of them sends the keywords it finds into the shared list, which is then cleaned and grouped.