A video ranks for more than its title keyword. Descriptions, chapters, captions and tags all carry secondary phrases that catch adjacent searches — but only if you know which phrases are actually adjacent. The Related Keyword Finder expands a seed keyword into 50+ semantically related terms, groups them by intent, scores each for opportunity, and maps the clusters so you can see the topic's shape.
This guide explains semantic relatedness versus string similarity, how to use intent groups, and how to turn the cluster map into a content series.
Semantic, not string
"Home workout" is string-related to "home workout plan" and semantically related to "no equipment exercises", "apartment-friendly cardio" and "resistance bands vs dumbbells". The second group is where the untapped searches are. The finder uses topic understanding rather than word matching, which is why the list includes phrases that do not contain the seed at all.
Intent groups
Informational terms go in descriptions and chapters of tutorial videos. Comparative terms ("X vs Y", "best X for Y") are usually their own videos. Transactional terms ("buy", "cheap", "deal") belong in review videos and affiliate content. Grouping by intent means you do not waste a comparative keyword as a tag on a tutorial where it will never rank.
From cluster map to series
The cluster map shows five to eight sub-topics. Each cluster with three or more high-opportunity terms is a video; the seed term is the pillar video that links to them all. That is a topical-authority structure YouTube rewards, built from one keyword.
Tips
- Seed with a two-word topic for the broadest useful map.
- Put the top three informational terms from the same cluster into one video's chapters.
- Run the top comparative term through the Keyword Difficulty Checker before making it a standalone video.
Common mistakes
- Using all 50 as tags — Tags are capped at 500 characters and irrelevant tags hurt classification.
- Mixing intents in one video — A tutorial that also tries to be a buying guide ranks for neither.
- Ignoring low-volume terms in a strong cluster — Together they add up; individually they look small.
Worked example
Input: "home workout"
What the report returned: 58 terms in 7 clusters. Largest: "no equipment" (14 terms). Highest opportunity: "home workout for small apartment no jumping" (74, informational). Comparative cluster: "resistance bands vs dumbbells at home" (68). Series: 1 pillar + 6 cluster videos mapped.
Frequently asked questions
How are the terms generated?
From topic modelling across autocomplete, related searches and video metadata in the niche, then de-duplicated and scored.
Are these different from the Long-Tail Keyword Builder?
Yes. This finds related topics (breadth). Long-Tail finds modifiers on one topic (depth). Use this first to pick the cluster, then Long-Tail on the cluster you chose.
How many should I use per video?
One primary in the title, two to three secondaries in the description and chapters, and up to ten in tags.
Can I export the cluster map?
Yes; it copies as a nested list.