LA.
All case studiesNext: Reel

Case study

Goodreads Redesign.

Rethinking how readers track and discover books — a focused redesign exercise.

Goodreads Redesign — hero
Year
2024
Type
App redesign
Role
Solo redesign
Timeline
3 weeks
Tools
Figma

The problem

Goodreads is loved for community but cluttered for everyday flows.

The outcome

Practiced restraint — knowing what NOT to change.

Highlights

  • Goodreads Redesign — My Books — current read leads, FAB to add

    My Books — current read leads, FAB to add

  • Goodreads Redesign — Social — like and comment inline

    Social — like and comment inline

  • Goodreads Redesign — Explore — recommendations and categories

    Explore — recommendations and categories

  • Goodreads Redesign — Review — half stars, quotes, notes

    Review — half stars, quotes, notes

The problem.

Goodreads is a beloved product because of its community — reviews, ratings, friend activity. But the core 'add a book, track progress, discover next' flows feel buried under a decade of feature accretion. The brief: respect what makes it loved, but make the everyday flows easier.

The approach.

  • Step 01

    Listening to long-time users.

    Goodreads is loved, but tolerated.

    Pulled real reviews and forum threads to map what current users love and what they put up with. The community kept surfacing as the thing people stay for; the everyday flows kept surfacing as the thing they tolerate.

    Users find that

    • Weird navigation
    • Not enough customizability — custom lists
    • Important tags are missing or not highlighted
    • Not enough insight into your reading habits
    • Things like quotes are not displayed well
    • Only discrete 1–5 ratings — no half stars
    • Home screen too cluttered with irrelevant content
  • Step 02

    What people are trying to do.

    Track, discover, review — the unglamorous core.

    Mapped what users were trying to do and where they stalled. Tracking books, discovering new ones, and writing reviews dominated the list. Those flows became the targets.

    As a user I want to

    • Track the books that I have read
    • Review and rate the books that I have read
    • Read reviews from other people
      • Prioritize reviews from friends
    • Discover new books I might want to read
      • Add and manage a want-to-read list
  • Step 03

    Where to focus, where to leave alone.

    Constraint, not reinvention.

    Community surfaces stay mostly intact. My Books, Social, Explore, and Review get rebuilt from the user's task — not the existing IA. Half-star ratings, a FAB to add a book, the current read promoted, inline social interactions, and notes-with-quotes on reviews — each tied to a specific pain. I also experimented with the nav itself: the most-used pages live on the right of the bar instead of the left, so the thumb has the shortest path to the things people actually do.

The hardest part of a redesign is what NOT to change.

The solution.

Goodreads Redesign — Before — My Books, current
Before — My Books, current
Goodreads Redesign — After — current read leads, FAB to add, TBR and recent one tap away
After — current read leads, FAB to add, TBR and recent one tap away
Goodreads Redesign — Before — Social, current
Before — Social, current
Goodreads Redesign — After — inline like + comment, add the book straight from the feed
After — inline like + comment, add the book straight from the feed
Goodreads Redesign — Home — what you're reading, who you follow
Home — what you're reading, who you follow
Goodreads Redesign — Review — half stars, plus quotes and notes
Review — half stars, plus quotes and notes
Goodreads Redesign — Explore — by recommendation
Explore — by recommendation
Goodreads Redesign — Explore — by category
Explore — by category

The outcome.

An exercise in designing within constraints rather than against them. Goodreads' community is its real product — design choices that ignored that got rejected by my own gut before they were worth testing. The everyday flows, by contrast, had room to move: every change tied back to a specific pain point users had already articulated.

What I’d do differently

I'd run real-user tests with longtime Goodreads users to validate that the IA shifts didn't disorient them — and pressure-test the half-star rating decision, since some readers may have built a personal scoring rhythm around the existing five whole stars.

Next case study

Reel.