Skip to content

Chapter 5 of 7

Distribution Surfaces and Platform-Specific Optimization

Each platform has multiple distribution surfaces where a piece of content can appear. On Instagram, your video may be surfaced in the Reels feed, Explore, suggested posts, hashtag pages, or the profile grid. On Facebook, the same piece may travel through the Reels feed, News Feed, Pages, Groups, Shares, or suggested content modules. The choice of surface affects who sees the content and how it is weighted, so creators benefit from understanding which modules are currently prioritized by the platform's ranking.

Packaging decisions interact with these surfaces. Hashtags can help with categorization and lightweight discovery, but ranking is usually dominated by watch-time and engagement signals, which is why hashtag impact is often limited compared to retention. Caption context matters because the text adds topic cues and clarity, improving both viewer comprehension and distribution to the right audience. Sound-on versus sound-off optimization acknowledges that many viewers watch silently, so videos should work without audio through captions and visuals while still rewarding sound-on viewers with clear audio and music.

Platforms also reward behaviors that improve whole-session metrics. Session starts count heavily because content that pulls users into longer sessions of multiple videos is associated with higher overall value. To balance different content types, platforms monitor feed cannibalization risk: too many Reels can crowd out other formats, prompting deliberate balancing across surfaces in service of strategic goals and user satisfaction.

Because small UI changes can shift engagement, retention, and ad revenue substantially, platforms experiment heavily with feed layouts, running tests that identify which adjustments actually move target metrics.

All chapters
  1. 1How Feed Recommendation Systems Work
  2. 2Ranking Signals and Their Tradeoffs
  3. 3When Optimization Goes Wrong
  4. 4Creator Growth and Content Design
  5. 5Distribution Surfaces and Platform-Specific Optimization
  6. 6Content Policy, Integrity, and Manipulation
  7. 7Trends, Identity, and Practical Strategies

Drill it

Reading is not remembering. These come from the Social Media Scrolling Virality Addon 60 V2 deck:

Q

SM: What is a “candidate set” in feed ranking?

The shortlist of posts/videos retrieved for you before final ranking decides what gets shown next.

Q

SM: What is “retrieval” in recommendation systems (simple)?

The step that quickly finds potentially relevant items from a huge catalog before more expensive ranking.

Q

SM: Why do platforms use multiple stages (retrieval → ranking)?

To scale: retrieval narrows billions of items to thousands; ranking then scores the shortlist more precisely.

Q

SM: What is a “feature” in ML ranking models?

An input variable used to predict outcomes (e.g., your recent watches, video length, creator relationship).