How to Find Chinese Web Novels Worth Reading (Without Speaking Chinese)
The problem was never that there is nothing to read. It is that there are two million books and no way to ask a question about them.
Most advice about finding Chinese web novels is a list of somebody's favourites. That is fine once, and useless the second time, because a list of ten books does not help you find the eleventh.
This is about method. Seven ways to interrogate a two-million-book catalogue, roughly in order of how much they reward the effort.
1. Start from a ranking board, not a search box
Search assumes you already know what you are looking for. A ranking assumes only that you want something good.
The best single entry point in Chinese web fiction is Qidian's monthly vote board. Monthly votes are bought with money and reset every month, so the board answers "what are paying readers spending on right now" rather than "what got a lot of clicks once". Fanqie's equivalent is Most Read, which measures a free action and so sets a lower bar, but is still the honest answer to what its enormous readership is actually reading.
The refinement that makes this much better: use the category slice. Overall boards are dominated by whatever is biggest, and in Qidian's case that is xuanhuan. If you want cultivation, go to the cultivation board. If you want mystery, go to that one. A book that is 60th overall and 4th in its category is a much stronger recommendation than the reverse.
Which board measures what is written up here.
2. Filter by trope, then sort by evidence
This is the highest-yield technique on this site and almost nobody arrives knowing it exists.
Chinese readers pick books by trope tag, not by genre, and those tags are structured data. So you can ask precise questions. Combine a tag with a sort:
The filter says what kind of book. The sort says which of them people stayed with. Neither alone is much use; together they are the closest thing to a good recommendation engine.
You can also use several tags at once in "all" mode, which is how you find the specific book you are imagining rather than a category of them. And the exclude filter matters as much as the include one: if harem is a hard no, exclude 多女主 and it is gone from every query you run afterwards.
The full tag glossary is here.
3. Read booklists
书单 (shudan) are reader-curated collections, and they are the best discovery mechanism these platforms have, because a person with taste assembled each one around a theme.
There are around 97,000 Qidian booklists and 14,000 Fanqie ones indexed here, and they average well over a hundred books each. A list titled something like "cultivation novels where the protagonist is genuinely clever" is doing work that no filter can do, because the criterion is a judgement.
The trick is to sort them by when the list was last updated rather than by how many people have seen it. A popular list from three years ago is a museum. There is a longer post on why booklists work so well.
4. Filter for finished books
The most common way to have a bad time with web novels is to invest four hundred chapters in something that stopped updating in 2022.
Two filters solve this. Status: completed means the author declared the book finished. A word-count range lets you pick something you can realistically finish: two million Chinese characters is a long but achievable book, ten million is a commitment measured in months.
The reverse query is also useful. If you want something live to follow week by week, Qidian's Most Updated board ranks purely by recent chapter output, which is the cleanest way to find books that are definitely still being written.
5. Use the right platform for the genre
The two catalogues are shaped so differently that picking the wrong one wastes an afternoon.
Qidian has about 71,000 cultivation titles. Fanqie has about 1,900. Fanqie has about 260,000 modern romance titles, roughly four times Qidian's. Neither platform is better; they are good at different things.
Rough allocation:
- Xianxia, xuanhuan, sci-fi, gaming, historical epics → Qidian
- Modern romance, urban, contemporary, ancient romance → Fanqie
- Anything trope-first → both, and compare
6. Follow the recommendation graph
Every book page here carries the books the platform itself associates with it, plus the booklists that book appears in. That second one is the underrated path.
If you find one book you like, look at which booklists contain it. Those lists were assembled by people who liked it too, and everything else on the list is a recommendation from someone whose taste you have just verified against one data point. It is a far better signal than an algorithmic "readers also liked".
7. Sort by durability, not by heat
Every platform's default sort is some version of "what is hot". Hot is a poor proxy for good, because it is dominated by whatever launched recently with a marketing push.
On Fanqie, all-time reads is the fix: it ranks by lifetime readership rather than current activity, which surfaces the books that were big and stayed big rather than the ones that are trending this week. Combine it with a tag and you have the durable hits of a specific trope, which is a query the Fanqie app itself cannot answer.
On Qidian, the equivalent is collections, which are cumulative, and especially VIP collections, which only count readers who had already paid before they committed.
Reading the English
Everything on this site is machine-translated with the original Chinese kept beside it, which is worth understanding rather than working around.
Titles translate badly, because they are often idioms or wordplay. Synopses translate reasonably well and are usually enough to decide from. Tags translate literally, which is why some read oddly: 魂穿 comes out as "Soul Penetration" and it means soul transmigration.
Treat the Chinese as the source of truth and the English as an index. That is what it is for: the point is not to read the book here, it is to be able to find the book at all, then follow the link to the platform that publishes it.
A starting sequence, if you want one
- Pick a genre from the rankings and look at its monthly-vote board.
- Open two or three books whose synopses appeal, and note which tags they share.
- Filter the catalogue by those tags, sorted by votes or by all-time reads.
- Find the booklists containing your favourite of those, and mine them.
- Exclude the two or three tags you have discovered you dislike, permanently.
That loop gets sharper every time you run it, which is more than can be said for anyone's list of ten favourites.