Skip to content

Discovery & Search

How users find books, authors, and friends in Logbook.

Overview

Discovery in Logbook prioritizes friend recommendations, but adds algorithmic suggestions to make the experience more tailored. The discover feed surfaces what friends are reading and loving, making book discovery feel personal and trustworthy.

Beta Launch

  • Search books, authors, people (your friends)
  • Featured books on discover page
  • Friend recommendations ("Friends loved these")
  • Trending books in your network

Post-Beta

  • Personalized recommendations (ML-based)
  • Similar books algorithm
  • Author follow
  • New releases from followed authors
  • Reading list suggestions
  • Seasonal/themed/genre collections

User Stories

  • As a user, I want to search for books by title or author so I can find specific books
  • As a user, I want to see what books my friends loved so I can discover new reads
  • As a user, I want to see trending books in my network so I know what's popular
  • As a user, I want to filter search results so I can find exactly what I'm looking for

Single search bar that finds:

  • Books (by title, author, ISBN w/ barcode scan in the future)
  • Authors
  • People (users)

Search Results

Results are tabbed: All | Books | Authors | People

Each result type has a distinct card style:

  • Books: Cover thumbnail, title, author, rating, friend count
  • Authors: Avatar, name, genre, book count
  • People: Avatar, name, handle, books read, mutual friends

Discover Feed

The discover page shows:

  1. Currently Reading - User's active books with progress
  2. Friends Loved These - Books rated 4+ stars by friends
  3. Trending in Network - Most added/finished books recently
  4. Because You Read X - Similar to books user enjoyed (post-beta)

Recommendation Philosophy

"Friends first, algorithms assist"

We prioritize friend recommendations over pure algorithmic suggestions - but we're not anti-algorithm. The reality is:

  1. Early users have sparse social graphs - New users won't have many friends on the platform yet, so we need algorithmic content to fill the gap
  2. Algorithms surface what friends might miss - Even with a strong friend network, algorithms can surface hidden gems
  3. The goal is authenticity - Algorithmic recommendations should feel like helpful suggestions, not manipulative engagement bait

How We Balance It

Content TypePriorityWhen to Show
Friend activity (reading, reviews)HighestAlways prominent
Books friends lovedHighPrimary recommendations
Trending in your networkHighSocial proof
Algorithmically similar booksMediumAugment when social content is sparse
Popular/trending globallyLowFallback for new users

Cold Start Strategy

For new users with few friends:

  1. Show popular books in their preferred genres (from onboarding)
  2. Surface "readers like you also enjoyed" recommendations
  3. Prompt to find friends / import contacts
  4. Gradually shift to friend-based content as social graph grows

The key: be transparent. Label algorithmic recommendations differently from friend recommendations so users know the source.

Built with VitePress