Feed Ranking Fundamentals
You'll learn why modern social media feeds show posts by relevance rather than purely by time, and the core principles behind ranking algorithms.
Why Ranking Matters
Imagine opening your social media feed to find 10,000 posts from the past day, sorted newest-first. You'd scroll through hundreds of mediocre posts before finding content you actually care about. Feed ranking solves this by predicting and displaying the posts you're most likely to engage with, regardless of when they were published.
Think of it like a personalized newspaper editor: instead of printing every article chronologically, the editor highlights stories they think you'll find most interesting on the front page.
Core Ranking Principles
At its heart, a ranking algorithm assigns each post a relevance score based on multiple signals:
- Recency: Newer posts typically rank higher (time decay)
- Engagement signals: Likes, comments, shares from your network
- Author relationship: Posts from close friends vs distant connections
- Content type preferences: Do you interact more with videos, photos, or text?
- User behavior patterns: What you've engaged with historically
The system sorts posts by these scores before delivering your timeline, showing highest-ranked content first.
Pull vs Push Impact
Ranking naturally fits fanout-on-read (pull) models since you compute scores at read-time with fresh data. In fanout-on-write (push) models, you pre-compute scores during fanout, but they may become stale. Hybrid systems often push posts unranked, then rank them when the user requests their feed.
Key Takeaway: Feed ranking algorithms transform chronological timelines into personalized experiences by scoring and sorting posts based on predicted relevance, balancing recency with engagement signals and user preferences.