A better way to discover fiction
Reading fiction is one of life’s great pleasures — and it’s good for us too. Even short periods of reading have been shown to reduce stress and improve mental wellbeing.
And yet, reading for pleasure is in decline. In the US alone, the number of adults who read regularly has fallen sharply over the past two decades.
There are many reasons for this, but we believe one stands out: choosing what to read has become surprisingly hard.
Bookstores and libraries now offer thousands of titles, with thousands more published every month. Faced with so much choice, many people simply don’t know where to begin — and often end up choosing nothing at all.
Recommendance exists to change that.
From recommendation to compatibility
Most book recommendation systems focus on popularity, sales, or broad categories. They’re often useful — but they rarely feel personal.
Recommendance takes a different approach.
Instead of asking “What books are popular?” we ask: “What kind of reader are you?”
The idea is inspired by online dating. Dating apps don’t try to find the best person — they try to find the right person for you. Recommendance does the same for books.
How it works
Recommendance matches readers and books using a shared framework.
Readers answer a short quiz designed to capture how they like stories to feel — for example, whether they prefer fast-paced plots or slow burners, lighter reads or weightier themes.
From this, each reader receives a concise Recommendance code that represents their reading preferences. The system is deliberately simple to use, but highly expressive: hundreds of thousands of distinct reader profiles can be represented.
Books are assessed using the same framework. Since we can’t ask books to take a quiz, we use AI to analyse a wide range of reviews, descriptions, and commentary to understand how each book reads in practice.
Once both sides are described in the same language, Recommendance can calculate compatibility — how well a particular book is likely to suit a particular reader.
A system in progress
Recommendance is an independent project, and it’s still evolving.
Right now, the system is available through a collection of focused mini-apps, each built around a specific, curated list — from literary classics to bestselling fiction across different decades.
We’re also running a live trial of Recommendance in an independent bookstore, where real readers are using the system to explore books on the shelves. Early results are encouraging, and the trial is helping shape what Recommendance will become next.
Our goal is not to replace browsing, serendipity, or human recommendation — but to lower the barrier to getting started, especially for readers who feel overwhelmed by choice.
Where this is heading
We believe book discovery should feel inviting, not intimidating.
Recommendance is being developed with bookstores, libraries, and readers in mind — as a calm, reader-first layer that helps people find fiction they’re genuinely excited to read.
If you’re curious, the best way to understand Recommendance is simply to try it.