Shellf vs StoryGraph

I’ll give credit where it’s due: StoryGraph has great reading stats - it's one of their most cited loved features. Mood tracking, pace analysis, page-count breakdowns, genre distribution charts, and their year-in-review reports are impressive and well-received. Nadia Odunayo’s team built something that makes data-driven readers very happy. There’s a reason they won an Apple App Store Award in 2025. My main issue is I find the UX not particularly intuitive.

I would argue that Shellf’s stats are different but on-par — books read, pages read, monthly trends, genre breakdowns, fiction vs non-fiction splits — but they’re not trying to compete with StoryGraph on raw quantitative depth. Where Shellf differentiates is with AI-generated reader insights through Shellf Plus: connections between your favourites, how your taste has evolved, your taste fingerprint, and reader archetype. It’s less about charting every dimension of every book you’ve read and more about understanding what your reading patterns reveal about you as a reader.

The honest assessment: if you love spreadsheets and charts and want to slice your reading data twelve different ways, StoryGraph is the clear winner. If you want an AI that looks at your library and tells you something you didn’t know about your own taste, that’s what Shellf is building. They’re genuinely different approaches, and I think there’s room for both.

One criticism I’ve seen of StoryGraph: their auto-generated mood and pacing data is frequently inaccurate. It’s pulled from a neural network rather than human input, and readers regularly report that books are tagged with moods that don’t match the actual experience. When your stats are built on inaccurate metadata, the beautiful charts can be misleading.

StoryGraph’s recommendation engine — as far as I can tell — uses a neural network trained on mood, theme, and genre data. It’s a step above Goodreads’ basic collaborative filtering, and for readers who know exactly what mood they’re in (“I want something dark and atmospheric”), it can surface reasonable options. But many users report the recommendations feel generic — they’re filtered by category rather than tuned to individual taste.

Shellf takes a fundamentally different approach. Instead of categorising books by mood labels, Shellf maps every book into a high-dimensional taste space using embeddings, then layers LLM reasoning on top. It doesn’t just know you like fantasy; it learns that you gravitate toward unreliable narrators, slow-burn character development, and morally grey protagonists. The recommendations come with explanations for why a specific book might resonate with you, based on patterns across your entire library.

I’m biased here — recommendations are the core reason I built Shellf. But I genuinely believe the embeddings + LLM approach produces better results than mood-based filtering once you have 30+ rated books. The tradeoff is that Shellf’s engine needs more data to work with. With fewer than 15 rated books, StoryGraph’s broader approach might actually be more useful.

StoryGraph also introduced AI-generated book descriptions, which have been controversial. Many users felt they were inaccurate or spoiler-prone. It’s a reminder that AI in book discovery is hard to get right, and I take that seriously with Shellf’s approach.

This is genuinely subjective territory, and I want to be fair. StoryGraph has a cleaner, more modern interface than Goodreads — that’s a low bar, but they clear it comfortably. The web app works well for readers who prefer desktop. The mobile app exists on both iOS and Android. It’s a capable, functional experience.

That said, I’ve seen consistent feedback that StoryGraph’s navigation can feel buried. Settings are sprawling. The mobile app doesn’t always feel native. Book sorting is limited — no secondary sort, no cover view. Performance issues and occasional outages crop up. Five million signups means real infrastructure pressure, and it shows at times.

My own personal experience is that I find Storygraph's UX not particularly intuitive, and not delightful. The stats screen is a bit hard to get to grips with and the book detail screen is cluttered.

Shellf was designed with a different philosophy: dark-first, opinionated, and focused. The bookshelf visual metaphor gives your library a tangible feel. Custom shelves use tag-based rules with any/all matching. Half-star ratings. Notes with voice dictation. DNF tracking with reasons. The tradeoff is that Shellf is mobile-only (Android for now), so if you want a desktop experience, StoryGraph wins by default.

I think the honest answer is: StoryGraph is more accessible (web + iOS + Android), and Shellf is more polished on the single platform it supports. Whether that matters depends on how and where you read.

StoryGraph’s free tier is genuinely generous. You get the core tracking, basic stats, content warnings, and recommendations without paying a penny. StoryGraph Plus at $49.99/year adds advanced stat filters, custom graphs, and some extra features. Many users have told me they don’t see enough value in Plus to justify the price — the free tier covers most of what they need.

Shellf’s free plan includes unlimited book tracking, full library organisation, reading stats, and 100 AI recommendation credits — enough for months of discovery for most readers. Shellf Plus at $18/year unlocks 2,000 AI recommendation credits, advanced reader insights, and priority features.

The math is simple: Shellf Plus costs 64% less than StoryGraph Plus. At $18/year — less than $1.50/month — it’s the cheapest premium tier of any book tracking app on the market. I’ve deliberately kept the price at a ‘no brainer’ level. My long-term goal is actually to make everything free, funded by affiliate revenue from book purchases made through the app. The Plus tier is primarily a safeguard against AI credit over-consumption while the business model matures.

If you never plan to pay for a book tracker, both free tiers are strong. But if you’re considering a premium upgrade, the value gap is significant.

This is an area where StoryGraph genuinely leads the entire industry. It's a great feature. Their crowdsourced content warning system lets users tag books with specific triggers, rated by severity. For readers who need to avoid certain content — whether for trauma, personal preference, or any other reason — it’s invaluable. No other mainstream book tracker offers anything comparable.

Shellf doesn’t have content warnings. It’s on the roadmap, but I won’t pretend a roadmap item is a feature. If content warnings are important to your reading life, StoryGraph is the right choice today, full stop.

When Shellf does implement content warnings, I want to do it thoughtfully rather than rushing out a half-baked version. StoryGraph has set a high bar, and readers deserve that level of care. But that’s a future story, not a present one.

If you want to try Shellf, the migration is straightforward. Export your library as a CSV from StoryGraph, then import it into Shellf. Your books, ratings, and reading dates transfer. The whole process takes under two minutes.

Shellf also supports direct Goodreads import via OAuth if your library originally came from there. And for readers with spreadsheets or data from other apps, Shellf’s generic CSV import with column mapping handles most formats.

My honest recommendation: you don’t have to choose one. Many readers use StoryGraph for its stats and content warnings while using Shellf for AI-powered discovery and library organisation. The apps serve overlapping but distinct purposes. Try Shellf alongside StoryGraph, see which features matter more to your actual reading workflow, and let that guide your decision.