How to Export Raindrop.io Bookmarks (HTML, CSV or TXT) in 2026

Raindrop gives you more than a bookmarks file — collections, tags, notes, highlights — which also means its export has more than one route and more than one choice to get wrong. The paths below are current as of October 2026, per Raindrop’s own help docs.

The two export routes (they’re not the same thing)

Whole-account backup — the one people mean. Open Settings → Backups in the web app and click Create new. Raindrop builds the archive in the background and emails a download link when it’s ready. This is the route that covers everything — bookmarks, collections, tags, highlights, notes — and it is the route to run before any migration, cancellation or “just in case.” The one exclusion worth knowing: files you uploaded (PDFs, images) are stored apart from the bookmark data; Raindrop’s docs send you to All bookmarks → Export → ZIP to pull those originals separately.

Selective export. Inside a single collection, the Export button writes just that collection. Selecting individual items (checkboxes on web, or per-item tick) narrows it further. Same formats, smaller scope — the route for “I only need the Work folder” without dragging the whole account.

Both routes offer HTML, CSV and TXT. Pick accordingly:

• HTML — the Netscape bookmark file. Universal: every browser imports it, and so do most bookmark managers, including Marqly. Raindrop’s own documentation says it plainly: choose HTML unless you have a reason not to.
• CSV — structured rows. Better when you want to look at your data: sort by created-at, filter tags, count per collection. Raindrop’s CSV keeps URL, name, excerpt, tags and creation dates as columns.
• TXT — a bare list of links. Fine as a reading queue; poor as a backup.

What survives the round trip

HTML is the conservative choice for a reason, but know exactly what it carries and what it can’t:

What you saved	In the HTML export	Notes
URL, title, folder tree	✅	Nesting is preserved; the destination may flatten it
Tags	✅	Comma-separated on the bookmark line
Description / note per bookmark	✅	Lands in the destination’s notes/description field
Highlights	⚠️ partial	In the whole-account backup, not as first-class fields in the HTML
Files (uploaded PDFs/images)	❌ in HTML	Separate Export → ZIP from All bookmarks
Nested folders beyond the destination’s depth	⚠️	The importing tool’s model decides how deep the tree survives

The flatten behavior is the usual surprise. Raindrop’s tree can go arbitrarily deep; a tool whose data model is “collections of bookmarks” maps the first level and folds the rest — Marqly, for example, keeps the first two levels as boards and deeper paths appear in the title, so a save in Work/Clients/Acme imports under board Work as item Clients/Acme …. The link and tags are intact; the shape simplifies. If your Raindrop tree is shallow, you’ll never notice.

The safety-net question

If you’re on Raindrop Pro, Settings → Backups also does scheduled daily snapshots — up to 30 restore points, with optional delivery to Dropbox, Google Drive or OneDrive — so “I deleted a collection last Tuesday” is a restore, not an archaeology project. Free accounts have no scheduler; the manual backup is the only path, which makes “export before anything risky” a rule, not a suggestion.

And if the real reason you’re exporting is that you’d rather find things by meaning than by folder — semantic search and AI tagging on import are exactly the Marqly migration covers; the Raindrop Export Analyzer previews what your file actually holds before any of it moves. Honest limits, per our own measurements: import your Raindrop HTML (its JSON export isn’t the format our importer reads) and note that imports take the import date, not the original created_at.

The short version

Settings → Backups for everything; collection Export for part of it; HTML unless you want a spreadsheet. Uploaded files need the separate ZIP. Pro schedules the safety net — on free, you are the scheduler.