Semantic Clustering of Search Demand. Use when you need grouping by questions rather than word roots alone; apply this method to the specified inputs and verify the deliverable
Installs into .claude/skills of the current project.
Sponsors
SSkills DirectorySkills Directory
Know which skills are safe — weekly.
SSkills DirectorySkills Directory
Is that skill safe to install?
Are you the author of Mk 238?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/backtrue-mk-238)
---
name: mk-238
description: "Semantic Clustering of Search Demand. Use when you need grouping by questions rather than word roots alone; apply this method to the specified inputs and verify the deliverable"
---
# MK-238|Semantic Clustering of Search Demand
For a regular chat, upload `packs/mk-238.md`; it includes the shared rules, this method, and required references
## Purpose
This Skill delivers: **Grouping by questions rather than word roots alone**
Execution steps compiled by ThinkWithBlack for this method's stated purpose, 2026-10-21; the course outline and original sources are retained below
If the account cannot show all terms, do not call the data complete market demand; the same term can mean different things in different service contexts
## Required Inputs
Actual search terms, source/period, product capabilities and existing pages; retain measurement definitions for any performance data
## Method Steps
1. Label each query’s job, audience, conditions and intent before grouping by wording
2. Group terms when one answer/page can serve the same job; split shared roots with different uses and retain ambiguity
3. Name clusters, list member queries and reasons, and check omissions/duplicates; without actual terms, label the output as a planning hypothesis
## Calculation and Interpretation
Cluster by question and destination; aggregate performance only on comparable periods and definitions, without inventing scores
## Required Output Fields
| Order | Field | How to complete |
|---:|---|---|
| 1 | Original query and source | Use the supplied source, calculation, or explicit unknown |
| 2 | Job and intent | Use the supplied source, calculation, or explicit unknown |
| 3 | Cluster | Use the supplied source, calculation, or explicit unknown |
| 4 | Member terms and rationale | Use the supplied source, calculation, or explicit unknown |
| 5 | Split or ambiguity | Use the supplied source, calculation, or explicit unknown |
| 6 | Destination | Use the supplied source, calculation, or explicit unknown |
| 7 | Next step | Use the supplied source, calculation, or explicit unknown |
## Verification cases
**MK-238-A1 | “Apple” refers to fruit storage and phone repair**
Expected: Create separate clusters
**MK-238-A2 | Different wording asks how to replace the same model’s battery**
Expected: Group while preserving original queries
## Method Sources
Original file: AI × Google Ads 搜尋廣告實戰課|五堂課程大綱
Source location: Clustering
Source basis: Extended from a course outline
Original URL: https://docs.google.com/document/d/10HOKMeR1qM9_LXmc7pYGXJ6tbOMJtBuFAxcuCCAJZk4/edit
English translation of source excerpt retained in the master table:
> • Keyword Expansion and Keyword Clustering
### Additional Source Locations
Source verification record for the original course material
- Source code: GO
- Original location: Clustering
- Original outline evidence (English translation): • Keyword Expansion and Keyword Clustering