Topic Selection
Ch03 Topic Selection & Niche Strategy
Topic selection determines over 60% of your outcome. The wrong genre, no matter how well-produced, hits an algorithmic ceiling. The right genre — one that precisely triggers user emotional needs — can go viral even with rough production. This chapter teaches data-driven topic selection and differentiation strategy.
Five Core Niches Analyzed
Niche 1: Domineering CEO / Hidden Identity (Largest market, ~35%)
Core hook: Identity gap → hidden revelation → reversal. The protagonist is mistakenly seen as weak/poor, endures mistreatment, then reveals their true powerful identity, publicly humiliating those who looked down on them.
Formula: Underestimated → Endure → Reveal → Reversal. At least one "reveal" must happen within the first 3 episodes to retain viewers.
AI advantage: Relatively simple settings (offices, mansions, restaurants). Focus on character consistency — lead appearance must remain stable across episodes.
Niche 2: Sweet Romance (~25%)
Core hook: High-frequency sweet interactions + mild conflict + happy ending. Users watch for "sugar" — at least 1–2 sweet moments per episode. Better for traffic revenue and brand sponsorship than pay-per-episode monetization.
Niche 3: Mystery & Reversal (~15%)
Core hook: Information asymmetry + revelation. Highest completion rate of all niches. Requires the strongest scriptwriting skill. Formula: Set mystery → plant clues → fake reversal → real reversal → lingering effect.
Niche 4: Xianxia Fantasy (~15%) — AI's biggest advantage
Core hook: Waste-to-chosen-one arc + cultivation + romance. Traditional live-action production costs 3–5x more due to costumes and sets. AI image generation creates any ancient Chinese or fantasy setting at near-zero cost. This is where AI creators have the clearest competitive edge over traditional production.
Niche 5: Urban Uplift (~10%)
Core hook: Ordinary person gets power → crosses class boundaries → real-world pressure relief. Broadest audience base. Key formula: Oppressed → unexpected advantage → step-by-step counterattack → class leap.
Data-Driven Topic Selection Tools
-
Feigua Data (feigua.io): Track top 100 dramas by play count, rising trend signals (50%+ growth), comment word clouds for emotional trigger mapping.
-
Chanmama: Revenue data for paid drama platforms, top 50 revenue earners, thumbnail and opening-second analysis.
-
Douyin Hot Spot Tool: Capture trending topics and extract their emotional core (not the event itself) to layer into your planned genre.
Blue Ocean Discovery: Four Methods
-
Cross-niche fusion: Combine two trending niches (e.g., CEO drama + mystery thriller; xianxia + modern urban).
-
Geographic differentiation: Rural/small-town settings dominate Kuaishou with far lower competition than city-based drama.
-
Age segment gaps: Viewers 50+ are the fastest-growing paying demographic, almost completely underserved by current content.
-
AI-native genres: Cyberpunk, post-apocalyptic, space opera — genres too expensive for live action. AI drama's exclusive blue ocean.
Five High-CTR Title Structures
| Type | Formula | Emotional Trigger |
|---|---|---|
| Identity Reversal | [Appears weak] + "is actually" + [Super identity] | Surprise, contrast satisfaction |
| Number Shock | Number + time/context + shocking result | Curiosity, satisfaction |
| Suspense Cliffhanger | Question/ellipsis + high-intensity emotion word | Curiosity, anticipation |
| Confrontation Direct | [I] + direct action + [target] | Catharsis, identification |
| Result First (flashback) | State result first → then explain how | Satisfaction + curiosity about process |
[TIP] Chapter Action Checklist
- Use Feigua Data to survey the top 50 dramas in the last 30 days — tally genre distribution to confirm your target niche.
- Find 5 dramas in your niche with 5M+ plays; run competitor analysis using Claude (adapt the prompt template from this chapter).
- Draft 3 candidate topic directions (each with a different differentiation angle), write 3 title variants for each using the structures above.
- Test your 9 titles with 5 people across different age groups; use their "would you watch this?" responses to finalize your topic.