The disadvantage of planning is poor risk resistance, but its advantage is high efficiency, allowing tasks to be completed efficiently in a short time; while the advantage of evolution is strong risk resistance, enabling continuous improvement without specific goals, but its disadvantage is low efficiency and time-consuming! They are like the Yin and Yang of Chinese Tai Chi... Yin represents evolution, silently updating and iterating where others can't see, continuously growing and accumulating energy; Yang represents planning, instantly gathering all energy and bursting at one point when a specific short-term goal appears. —— Chapter 23 Evolution Strategy

1. Core logic of evolution strategy: Genetic variation + Natural selection, both are indispensable
2. Four-step cycle: MVP (initial state) → Natural selection (market feedback) → Variation (innovation) → Heredity (accumulating advantages)
3. Step 1: Launch initial state (MVP). Don't hold back a big release; build a minimum viable product around core pain points and enter the market quickly
4. Step 2: Natural selection. External scissors (market feedback/user reviews) + Internal scissors (horse race system/last-place elimination), screen with real feedback
5. Step 3: Variation. Recombinative innovation (combining excellent elements to create new species) or disruptive innovation (creating new features through internal "destruction")
6. Step 4: Heredity. Solidify the advantages that won in natural selection into the next generation and continue to accumulate
7. Complementary relationship between planning and evolution: Use planning to improve efficiency for high-certainty parts, and use evolution to test errors for high-uncertainty parts
8. The direction of evolution is not determined by oneself, but by the environment (market) — "The product you think is good may not be accepted by the market"

• Birth of WeChat: Version 1.0 only had address book and free text messages, a minimalist MVP launched → iterated based on user feedback → gradually added voice, Moments, payment and other features
• Xiaomi MIUI: First developed an optimized Android system instead of directly making mobile phones, faced user complaints directly through the community → iterated weekly → after MIUI matured, users actively requested to make mobile phones → Xiaomi phones emerged
• Tencent's horse race system: Before WeChat was launched, three teams (QQ team, Chengdu team, Guangzhou Research Institute) developed similar products simultaneously, competed internally, Zhang Xiaolong's team won, and others were disbanded
• Recombinative innovation of iPod: iTunes came from the transformation of SoundJam MP, hard disk from Toshiba, system from Pixo, battery from Sony — combination of excellent genes

• Startup/new product direction is unclear, unsure what users want
• Market environment changes too fast, plans can't keep up with changes
• Need to verify assumptions but have limited resources
• Organization or team is rigid and lacks vitality

• "Unsure what the market wants"
• "Let's get started first"
• "Plans can't keep up with changes"
• "How to verify ideas quickly?"
• "The team lacks innovation vitality"

• vs. systems-thinking: Systems thinking is an analysis tool (diagnosing system structure), while evolution strategy is an action framework (designing iteration cycles)
• vs. innovation-method: Evolution strategy focuses on the entire iteration cycle (including MVP, feedback, heredity), while innovation methods focus on specific techniques for the variation step
• vs. scientific-decision: Scientific decision-making is used for "choosing among known options", while evolution strategy is used for "the exploration stage where options have not yet appeared"

• Clarify the core pain point of target users (focus only on one most painful point)
• Develop a usable product/solution with minimal feature combinations, focusing on solving the pain point
• Internet products can be directly launched to run data; physical products need to be optimized to the current best before launch
• Completion standard: The product can be used by users in a real environment and can collect feedback data

• Launch the product to the real market and collect user feedback (likes/complaints/usage data)
• External scissors: Establish user feedback channels (community/customer service/data tracking), and reluctantly cut features that the team likes but users don't accept
• Internal scissors (optional): Set up internal competition mechanisms (horse race system/last-place elimination) to allow multiple solutions to compete simultaneously
• Completion standard: Collect at least one round of real market feedback and complete one iterative optimization based on it

• Based on the parts that won in natural selection, carry out variation and innovation (refer to innovation-method skill)
• Solidify the advantages verified by the market into the next generation of products (heredity)
• Start the next evolution cycle: Launch → Feedback → Variation → Heredity
• Completion standard: Complete at least one full cycle of "Launch → Feedback → Optimization", with quantifiable improvement indicators

• Projects with clear goals and clear paths (planning is more efficient, evolution wastes time)
• Fields with extremely low fault tolerance such as safety/medical care (cannot "launch first then iterate")
• Situations with extremely limited resources that cannot afford multiple rounds of trial and error

• MVP quality is too low: Treating semi-finished products as MVP, resulting in extremely poor user experience and even loss of initial users
• Only variation without heredity: Starting over with each innovation, not accumulating verified advantages
• No real feedback: Indulging in internal analysis and avoiding the cruel judgment of the market
• Excessive horse racing: Internal competition consumes too many resources, resulting in lower overall efficiency instead

• Lack of clear standards for "how to judge when to use planning and when to use evolution", which easily leads to an either-or situation in practice
• The horse race system is costly and unrealistic for small and medium-sized teams, but the effect of alternative solutions (such as last-place elimination) is quite different
• Lack of discussion on controlling the speed of evolution — iterating too slowly will be overtaken by competitors, while iterating too fast means each feedback is insufficient

• mirror-world (prerequisite): Evolution strategy is essentially a strategy to deal with the "probability domain" and "risk domain" in the mirror world. First, use the mirror world framework to judge the uncertainty level of events, then decide whether to use planning (high certainty) or evolution (high uncertainty).
• innovation-method (collaboration): The "variation" step of evolution strategy requires innovation methodologies to provide specific techniques for recombinative and disruptive innovation. Evolution strategy manages the entire iteration cycle, while innovation methods manage "how to change".
• compound-effect (comparison): Compound effect emphasizes deterministic growth of "finding the causal fulcrum and sticking to the milestone", while evolution strategy emphasizes addressing uncertainty through "letting the environment screen directions through trial and error". They are suitable for problems of different natures.