Bluesky Feed Analytics

[burningtree]

One of Bluesky's strongest weapons and appeal is that users can choose their own algorithms through custom feeds. Unlike traditional social platforms that force users into a single algorithmic experience, Bluesky embraces algorithmic choice and customization.

Making aggregate statistics from Bluesky's AppView infrastructure publicly accessible would give us valuable insights into how users interact with and adopt different feeds across the platform.

The Challenge

• No public statistics about feed adoption rates
• Lack of visibility into which feeds are actually being used
• No clear metrics about user engagement patterns with different feeds

Potential Solution

Making aggregate statistics from Bluesky AppView publicly accessible would:

• Provide transparency into feed usage patterns
• Help creators understand which types of feeds resonate with users
• Enable data-driven improvements to feed algorithms
• Foster better understanding of user preferences in algorithmic content curation

Possible Feed Analytics Metrics

Usage Metrics

• Number of active subscribers per feed
• Daily/monthly active users per feed
• Retention rates for custom feeds
• Feed switching patterns
• Time spent in different feeds

Engagement Metrics

• Post interaction rates within feeds
• Feed discovery patterns
• Popular feed combinations

Technical Metrics

• Feed loading times
• Cache hit rates
• Query performance
• Feed availability metrics
• Update frequency stats