10 places pirated copies of your book end up (and how to check them all)

Most authors find out by accident. A reader emails to ask why the PDF they downloaded has a scrambled table of contents, or a Google Alert fires on a domain you’ve never heard of, and suddenly your book has been sitting on a download site for eight months. If you want to know how to find out if my book is pirated before a stranger tells you, the answer is a checklist, not a tool. Ten places account for almost all of it, and you can work through them in an afternoon with a browser and one distinctive sentence from your own book.

The one search string you’ll reuse ten times

Pick a sentence from the middle of your book. Not the first line, which shows up in every retailer preview, and not a line with your title or your name in it. You want a sentence that exists nowhere except inside the full text: fifteen to twenty words, no proper nouns, no dialogue that could be a common phrase.

Put it in quotes. That quoted string is your fingerprint, and it’s the single check that separates a real leaked copy from a bookstore listing. Search engines index the text of uploaded PDFs, so a page that returns your fingerprint is serving the inside of your book, not a cover image and a buy button.

Keep three things open as you go: a notes file for URLs, a screenshot folder, and the date. You’ll need all three when you file.

1. Ebook libraries

Anna’s Archive, Z-Library, and Library Genesis are the highest-volume destinations for pirated books, and they mirror each other. A copy on one appears on the others within weeks. Search your title, your author name, and your ISBN on each. These sites move domains often, so search the name of the library rather than a bookmarked URL.

Removal routes differ between them and so does the odds of success. We cover that split in Z-Library vs Libgen vs Anna’s Archive.

2. Direct-download ebook sites

OceanofPDF and the cluster of sites built on the same template hand out EPUB and PDF copies with a two-click download. They rank in Google for “yourtitle pdf free download”, which is the point of them. Run that exact search with your title in place of the placeholder and look past the first three results, where the retailers sit.

These sites answer DMCA notices at a reasonable rate. Their clones and mirrors do not, which is why one takedown rarely closes the file.

3. File lockers and shared drives

Mega, Mediafire, and public Google Drive folders hold the copies that get passed around in private. Nobody links to them from an indexed page, so a normal search finds nothing.

Two checks work here. Search your fingerprint sentence with site:mega.nz and site:mediafire.com, then search your title alongside the words “drive folder” and “mega link”. The Google Drive hits surface when someone pastes the share link into a forum post that does get indexed.

4. Torrent indexes

Torrent sites carry books in bundles: an author’s complete works, a genre pack, a “2026 bestsellers” archive of four hundred titles. Search your author name rather than the title, because the bundle name rarely lists individual books.

The index itself hosts a magnet link, not your file, and many operators ignore notices on that basis. Document what you find and treat it as evidence of spread rather than a takedown target.

5. Document and study platforms

Scribd-style upload platforms and study-material sites take user uploads and put them behind a signup wall, which means Google indexes the text while the page looks legitimate. Course Hero is the one that comes up most for anything used in a classroom, including non-fiction and textbooks.

Search your fingerprint sentence here. The wall stops a human reader, not the crawler, so a quoted phrase from page 140 will surface the document.

6. Telegram and Discord

Telegram channels distribute ebooks at scale, and Discord servers trade them in pinned messages. Neither is indexed by Google, so your fingerprint search returns nothing.

Search inside Telegram for your title and your author name using the app’s own search, which reaches public channels. For Discord, search your title with site:disboard.org and check any server whose name includes “books”, “library”, or “archive”. Report to the platform, not the uploader.

7. Facebook groups and Reddit

Private Facebook groups run request-and-fulfil threads: a member names a book, another member posts a link or a file. Search Facebook for your title and filter to Groups. Reddit is more open, so your fingerprint search with site:reddit.com catches the threads where someone pasted a chunk of text.

Both platforms act on reports about specific posts. Neither will act on a report about a group.

8. Unauthorized retail listings

This one costs you money in a way the others don’t. Counterfeit paperbacks appear on Amazon, eBay, and AliExpress, printed from a scanned PDF and shipped to a buyer who believes they bought your book. On Etsy and Gumroad you’ll find “instant PDF download” listings of titles the seller never wrote.

Search your exact title on each marketplace and compare every listing against your own editions: publisher name, page count, cover version. A seller with no other books and a price under half of yours is worth a closer look.

9. Audiobook rips

Full audiobooks get uploaded to YouTube as single eight-hour videos with a static cover image, and to podcast feeds as chapter-per-episode series. Search YouTube for your title plus “full audiobook” and sort by upload date.

If you have an audiobook edition, this is the leak that takes the most revenue per copy, because the listener has no reason to buy one after finishing yours for free.

10. Scraper and “read online” sites

The last category is the strangest. Sites that host your full text chapter by chapter, wrapped in ads, often with the text run through a translation loop that leaves the prose slightly wrong. They exist for ad revenue.

Your fingerprint search catches the faithful ones. For the mangled versions, search a distinctive proper noun from your book together with a phrase like “read online free”.

Turning the list into a filing

A URL alone won’t get anything removed. For every hit, save the page URL rather than the domain, take a dated screenshot, and note which of your editions it copies. Ten minutes of this per hit is what turns a list into a set of notices a host will act on.

Where you send it depends on who’s holding the file. For a site with a real business behind it, find its designated agent in the Copyright Office DMCA agent directory and send there. For a platform, use its own form: YouTube’s copyright complaint form for audiobook rips, the marketplace’s infringement report for counterfeit listings. Submit the URL to Google’s legal removal request as well, which pulls the page out of search even while the host sits on your notice.

Our DMCA takedown notice templates cover the wording for each of those routes, and what to do when someone steals your content walks through the sequence in order.

How Clampdown helps

Running this checklist once tells you where you stand today. The problem is that it goes stale in a fortnight, because a removed file comes back under a new filename on a mirror you hadn’t found.

Clampdown runs the same ten checks as a continuous scan. You register a title once, we search the libraries, download sites, lockers, marketplaces, and upload platforms for copies, verify each match against your text before acting, file the notices and the search de-indexing, and keep watching so a re-upload gets caught without you noticing it first. Start with a free piracy audit to see the current list, or read the site-by-site removal guides for the hosts that come up most.

FAQ

How do I know a listing is a pirated copy and not a legitimate reseller? Compare it to your own editions. A real reseller lists your publisher, your ISBN, and a plausible price. A pirated listing tends to have a mismatched page count, a low-resolution cover, and a seller account with no history.

Does a copy on a torrent site mean my book is everywhere? It means your book is in a bundle, which is how it reaches the libraries and download sites next. Treat a torrent hit as a reason to check the other nine places rather than as the problem itself.

How often should I run this check? Quarterly for a backlist title, monthly in the three months after a launch. New releases draw the most uploads in their first weeks.

Will a takedown on one site remove the copy from the mirrors? No. Each host needs its own notice, and the libraries in particular mirror each other, so a file removed from one can be re-seeded from another within days.