MINY Strat
Thesis
The independent music creator economy is a $5B+ market with 60M+ active creators globally. These creators need strategic guidance but can't afford managers, consultants, or label support. MINY Strat packages expert-level music business strategy into an AI-powered conversational interface — available 24/7, in any language, at negligible marginal cost.
Product
MINY Strat is a strategy intelligence system with two interfaces:
- Telegram bot (@minystratbot) — conversational, meets artists where they already are
- Web app (skills.minyvinyl.com) — browse and explore the full strategy database
The system draws from a curated database of 1,597 strategies extracted from proprietary music business research, covering 12 action domains. Unlike generic AI chatbots, it:
- Searches a real database (not hallucinating)
- Remembers artist context across sessions
- Works in any language
- Limits output to 2-3 actionable items (not walls of text)
- Cites sources (strategy IDs traceable to original material)
Market
| Segment | Size | Pain Point |
|---|---|---|
| Independent artists | ~50M globally | No strategic guidance, overwhelmed by generic advice |
| Small managers | ~2M | Managing 1-5 artists, no research budget |
| Music educators | ~500K | Need structured curriculum resources |
| Labels (indie) | ~100K | Scouting and artist development at scale |
The advisor/consultant market charges $100-500/hour. A subscription bot at $5-15/month democratizes access to the same strategic thinking.
Business Model
Unit Economics
The cost structure is radically simple. The LLM runs at flash-lite pricing ($0.075/1M input tokens). The strategy database is static JSON loaded in memory. No vector database. No embeddings pipeline. No GPU inference — just keyword search with weighted scoring.
Competitive Landscape
| Competitor | Approach | Weakness |
|---|---|---|
| ChatGPT / Claude | General AI | Hallucinates music strategy, no curated database, no artist memory |
| Berklee Online, etc. | Static curriculum | $500-5,000, no personalization, no conversation |
| Management consultants | Human 1:1 | $100-500/hr, not accessible to independents |
| Spotify for Artists | Platform data | Data-only, no strategic interpretation |
| Bandzoogle / DistroKid | Content marketing | Generic articles, not personalized to artist situation |
MINY Strat's moat is the curated, structured strategy database combined with artist-specific context. The database is proprietary. The artist memory creates switching costs — the bot knows your situation, your history, what strategies you've already explored.
Traction
- Bot live on Telegram (@minystratbot)
- Web app live at skills.minyvinyl.com
- 1,597 strategies indexed across 12 domains
- Multi-language support (tested: English, Spanish, French)
- 6 commits shipped in one sprint (edge cases, memory, translation, response discipline)
No paying users yet. Product is in validation phase — testing with real artists to validate the interaction model before monetization.
Technical Differentiation
- No hallucination by design. The bot only returns strategies from the curated database. If there's no match, it says so honestly.
- Code-enforced guardrails. Critical UX rules are enforced in TypeScript, not just in prompt instructions. Lightweight LLMs follow prompt rules ~70% of the time.
- Zero-infrastructure AI. No vector DB, no embeddings, no GPU. A 892KB JSON file + keyword scoring outperforms semantic search for a domain-specific dataset.
- Language-agnostic. Queries auto-translate to English for search, responses render in the user's language. Cost: $0.
Team
Collective Win (collectivewinca) — building at the intersection of music, AI, and creator tools.
Ask
Product validation phase. Looking for:
- 50 independent artists to test the bot and provide feedback
- Partnership with a music distributor or creator platform for distribution
- $25M raise for content licensing, additional strategy databases, platform expansion, and hiring a full music industry advisory team
Risks
1. Strategy database depth
1,597 strategies is substantial but not infinite. Artists in niche genres or unusual markets may exhaust relevant results quickly.
2. LLM compliance
Flash-lite models are cheap but unreliable at following complex instructions. Upgrading to a more capable model increases cost 10-50x.
3. Content licensing
Strategies derived from proprietary source material may face IP challenges at scale. Phase 2 needs original content or explicit licensing agreements.
4. Distribution
Telegram is niche in the US music market. WhatsApp, Discord, or native app integration may be needed for growth.
Don't Submit. Apply.
Get access to the latest models and early releases through the MINY Insider Program.