Lesson 3: Keyword Research Fundamentals · Lesson 3.1

Build a Seed Keyword List

Start keyword research from problems, outcomes, jobs, features, and audience language.

Why this lesson matters

A weak seed list creates weak metadata because later research can only refine what you start with.

Core idea

The seed list is not the final keyword strategy. It is the raw material that later gets filtered by relevance, competition, and conversion fit.

Real-world example

A sleep tracker finds better keywords from user language

The product team says "recovery score," but users write reviews about "sleep quality" and "waking up less tired." Those are better starting points for research.

Why the example matters

The best seed terms often come from how users describe the problem, not how the team describes the feature.

Let's make it clearer

Gather seed terms from demand, not only from features

A strong seed list begins with the language of problems, outcomes, user roles, and alternative behaviors. Feature terms matter, but they should not dominate the first pass. If the seed list starts too close to internal product language, the final metadata often misses how users actually search and compare.

Students should treat the first seed list as research material, not as a draft metadata set. The goal is to capture the widest useful picture of demand before later scoring and compression happen.

Clean the list before it becomes noisy

Seed lists become messy fast because every new idea feels worth saving. That is why the raw list should be grouped early into problem terms, outcome terms, category terms, feature terms, and competitor terms. Structure prevents later scoring from turning into guesswork.

This grouping step also reveals gaps. If the list is full of features but empty on outcomes, the product promise is probably weak. If it is full of broad category terms but thin on user language, the team may still be thinking too much like builders instead of buyers.

Keep one bank for discovery, not for final decisions.

Tag where each term came from so later review is easier.

Remove duplicates and obvious low-fit terms before scoring.

Step-by-step framework

Step 1

Start with problems users want solved.

Step 2

Add outcome language users care about.

Step 3

Add tool and category language only after problem and outcome language exist.

Step 4

Group the list by use case before reviewing competitors.

Practical exercise

Write 25 seed keywords from problem, outcome, feature, audience, and competitor language buckets.

Key takeaways

Seed research should start from user demand.

Features alone are not enough.

Structure early makes later scoring easier.

Soft transition

Turn the seed list into a working bank

ASO Miner can help you organize competitor language and keyword candidates once the first seed list is ready.

Continue within this lesson

Next lesson in the academy

Competitor and Category Mining

Extract recurring language from competitor names, subtitles, screenshots, and review themes.