Feature Flag Best Practices

Table of contents

1. 1. Introduction
2. 2. The Moving Parts of a Feature-Flagging System
   1. Creating Separate Code Paths
3. 3. Best Practice #1: Maintain Flag Consistency
4. 4. Best Practice #2: Bridge the “Anonymous” to “Logged-In” Transition
5. 5. Best Practice #3: Make Flagging Decisions on the Server
   1. Performance
   2. Configuration Lag
   3. Security
   4. Implementation Complexity
6. 6. Best Practice #4: Incremental, Backward-Compatible Database Changes
   1. Code First
   2. Data First
   3. Big Bang
   4. Expand-Contract Migrations
   5. Duplicate Writes and Dark Reads
   6. Working with Databases in a Feature-Flagged World
7. 7. Best Practice #5: Implement Flagging Decisions Close to Business Logic
   1. A Rule of Thumb for Placing Flagging Decisions
8. 8. Best Practice #6: Scope Each Flag to a Single Team
9. 9. Best Practice #7: Consider Testability
10. 10. Best Practice #8: Have a Plan for Working with Flags at Scale
    1. Naming Your Flags
    2. Managing Your Flags
       1. Establishing a Retirement Plan
       2. Finding Zombie Flags
       3. Ownership and Change Control
       4. Flagging Metadata
11. 11. Best Practice #9: Build a Feedback Loop
    1. Correlating Changes with Effects
       1. Inference
       2. Causality
    2. Categories of Feedback
12. 12. Summary