Understanding Algorithms: How They Shape Your Digital Experience - Cheat Sheet

Algorithm Basics

• Definition: Sets of rules or instructions computers follow to solve problems
• Purpose: Process data to deliver personalized experiences, recommendations, and results
• Prevalence: Present in social media feeds, search results, recommendations, email filtering, pricing, and content moderation

How Personalization Algorithms Work

1. Data Collection Phase
• Explicit data: Information you deliberately provide
• Implicit data: Information inferred from your behavior
• Social data: Information from your connections
• Contextual data: Time, location, device type
2. Analysis and Prediction Phase
• Creating user profiles based on interests
• Identifying behavioral patterns
• Comparing to similar users
• Predicting engagement likelihood
3. Curation and Delivery Phase
• Ranking content by predicted relevance
• Filtering out “irrelevant” content
• Promoting content similar to previous engagements
• Testing new content types to refine understanding

Algorithm Business Objectives

• Maximizing engagement (time spent)
• Increasing conversion (purchases)
• Building habits (regular usage)
• Collecting more data
• Reducing operational costs

Platform-Specific Algorithms

• Facebook/Instagram: Prioritize interaction-generating content
• TikTok: Focus on watch time and completion rate
• Twitter: Balance chronology with engagement
• Google: Use hundreds of factors including relevance, authority, and user experience
• Netflix/YouTube: Predict what will keep you watching
• Spotify: Create personalized playlists based on listening patterns
• Amazon: Suggest products based on purchase history and similar customers

Algorithm Control Techniques

• Feed Training
• Be intentional about engagement (likes, comments, shares)
• Use “not interested” options to reject content
• Pause viewing history temporarily
• Periodically clear history to reset algorithms
• Create multiple accounts for different purposes
• Alternative Access Methods
• Use chronological feeds when available
• Subscribe directly via email or RSS
• Use third-party aggregation tools
• Search for specific content rather than browsing feeds
• Settings Adjustments
• Review privacy settings to limit data collection
• Adjust ad preferences
• Manage notification settings
• Explore platform “well-being” features

Algorithmic Audit Questions

1. What topics dominate my feeds?
2. What emotions do recommendations tend to evoke?
3. What perspectives seem overrepresented or underrepresented?
4. How much diversity of content do I see?
5. How often am I shown content that challenges my views?

Ethical Considerations

• Transparency: Right to know how algorithms make decisions
• Agency: User control over algorithmic systems
• Accountability: Responsibility when algorithms cause harm
• Diversity: Different impacts on different groups
• Attention economy: Fairness of attention monetization