What's actually in your X data archive

Your X (Twitter) data archive contains your posts, replies, likes, direct messages, profile history, follower and following lists, and the media you've uploaded, plus a surprising amount of ads data. It does not contain your bookmarks. That absence surprises almost everyone who requests an archive to rescue their saves, so let's settle it in the first paragraph: if bookmarks are what you're after, the X data archive won't help, and you'll have waited a day to find out.

Here's what's inside, how to get it, and what to do about the gaps.

How to request your archive

The request lives in your account settings, and the path is the same on web and mobile:

1. Open Settings and privacy, then Your account.
2. Choose "Download an archive of your data".
3. Re-enter your password, then verify with the email or phone code X sends you.
4. Tap Request archive, and wait.

The build isn't instant. Around 24 hours is typical, and bigger or older accounts can take longer; X pings you in-app and by email when the ZIP is ready.

The download link doesn't live forever, so grab the file promptly and store it somewhere durable.

What you'll find in the ZIP

Unzip it and you get two things: a browsable viewer and the raw data.

The viewer is an HTML file (named along the lines of "Your archive") that opens locally in your browser and lets you click through your own posting history offline. It's genuinely pleasant, and for most people it's the only part they'll open.

The data folder holds the substance, mostly as JavaScript/JSON files. The highlights:

• Posts and replies, full text with timestamps and engagement counts, plus your uploaded images and video as actual files. Media is why archives from photo-heavy accounts run to gigabytes.
• Likes, as a list of liked posts, in a trimmed format with post IDs and text snippets.
• Direct messages, complete conversations, including ones you'd forgotten you had. Treat the ZIP as sensitive; it's your DM history in plain files.
• Account and profile data: past display names, bios, email and phone records, login-relevant metadata.
• Followers and following, as account ID lists.
• Ads data: ad impressions and the interest and audience categories X's ad system has assigned you, which is worth a bemused skim.

For posts, likes, and DMs, this is a genuinely complete export, and I recommend requesting one yearly regardless of any bookmark concerns. It's the only offline copy of your side of X you can get.

Reading the raw files, if you want to

One quirk trips up anyone who tries to process the data: the files in the data folder are JavaScript, not plain JSON. Each one wraps its records in an assignment like window.YTD.tweets.part0 = [...], because the bundled HTML viewer loads them as scripts. To use them elsewhere, open the file, delete everything before the first [, and save it with a .json extension; every spreadsheet tool and scripting language takes it from there. Large accounts get their posts split across several numbered part files, so check for a part1 before assuming you have everything.

Two practical notes while you're in there. Sizes run from a few megabytes for quiet accounts to multiple gigabytes for media-heavy ones, so don't request it on hotel wifi. And since the ZIP holds your full DM history and account metadata in readable files, store it like a password export, not like a photo album.

What's missing, starting with bookmarks

Bookmarks are the headline omission: not present, and they never have been, which multiple independent write-ups of the archive's contents confirm.

X has never publicly explained the choice. The pattern behind the archive seems to be "data you created": posts you wrote, messages you sent, media you uploaded. A bookmark is a private pointer at someone else's content, and X appears to treat the pointed-at posts as not yours to take, the same reason your Likes arrive as a trimmed list rather than full copies of other people's posts. That's my reading, not an official statement.

Also absent: other people's content generally (a bookmarked or liked post that gets deleted is not preserved anywhere in your archive), your bookmark folder structure if you pay for Premium, and anything about how the algorithm ranks you.

Filling the bookmarks gap

Since the archive won't cover them, bookmarks need their own pipeline, and the tools are mature:

• A one-time copy: free browser extensions like XSaved or Twillot read your bookmarks page and produce CSV or JSON files; the walkthrough is in how to export your X bookmarks.
• An ongoing copy: sync services keep the file current automatically, compared in how to back up your X bookmarks.
• If saves have already gone missing: work out which failure you're looking at in why your X bookmarks disappeared.

One structural note for API-based tools: X's API returns at most your 800 most recent bookmarks, so tools built on it, Bookmarx included, see a sliding window rather than your full history. Details and sources are in the X bookmark limit, explained, and the full series is on the X bookmark guides hub.

My own habit, for what it's worth: archive request every January for posts and DMs, an extension export of bookmarks quarterly, and Bookmarx running in between, since it syncs and summarizes each new bookmark from the moment you connect and keeps its own copy of the text even if the original post later disappears. The archive covers what you've said; something else has to cover what you've saved.

Frequently asked questions

Are bookmarks included in the X data archive?

No. The archive covers your posts, likes, DMs, profile data, and follower lists, but bookmarks have never been part of it. Saving bookmarks requires a browser extension or a third-party sync service instead.

How long does the X archive take to arrive?

Plan on around 24 hours, sometimes longer for large or old accounts. X notifies you in the app and by email when it's ready, and you then have a limited window to download the ZIP before the link expires.

Does the archive include my photos and videos?

Yes, media you posted is included as files inside the ZIP, which is a big part of why archives from media-heavy accounts run to gigabytes. Media from other people's posts, including posts you bookmarked or liked, is not yours and isn't included.

Can I open the archive without technical skills?

Yes. The ZIP ships with an HTML file, usually named Your archive.html, that opens in any browser and gives you a local, clickable view of your posts and data. The underlying data files are there too if you want to process them.