How to Export Journey App Entries and What Stays Behind

2026-09-21 · Yesterself

An export from a cloud-synced journal is not a copy of your journal. It is a copy of whatever was on the device you pressed the button on, plus whatever that device managed to fetch while the export was running. On a single-device app those are the same thing. On a cross-platform journal that moves between an Android phone, a laptop browser and maybe a second tablet, they routinely are not, and the difference is invisible until the archive is the only copy left.

That gap is the subject here. Not where the export button is, which moves between versions anyway, but what the downloaded bundle actually contains, what only ever existed on the server, and what is worth capturing before you stop paying for the account that holds it.

If you want the general mechanics of exporting and verifying a journal archive, the step-by-step version for a single app covers the counting. This piece is about the sync problem specifically.

Your device holds a partial copy

Cross-platform journals keep the authoritative copy on a server and give each device a working mirror. Mirrors are allowed to be incomplete, and usually are, for reasons that made sense at the time:

• Storage optimization. Full-resolution photos and video live in the cloud, and the phone keeps a small preview until you open the entry. An export built from the device may collect the previews.
• Reinstalls and new devices. A phone you set up in March may have back-filled only recent entries, or may pull older ones lazily as you scroll.
• Platform gaps. Something written in the web version last week may not have landed on the tablet you have not opened since spring.
• Journals or notebooks that are switched off. If the app lets you choose which collections sync to which device, the ones you unchecked are not there.

None of this is a defect. It becomes one at the moment you export from the thinnest mirror you own and treat the result as your archive.

So the first step is not exporting. It is this: open the app on every device you use, let each one sync to completion on wifi and mains power, and only then decide which device to export from. Pick the one with the most complete local copy, which is usually a desktop or laptop rather than a phone, both because storage is less contested and because you get a folder you can open and inspect instead of a zip trapped behind a share sheet.

What comes down, and what does not

Once you have the bundle, sort its contents into three piles in your head.

Entry data. The words, the dates, and usually the tags. This is the part that exports well and the part you most care about. In a structured export it will be a JSON file or a folder of text files.

Attachments. Photos, audio, video, PDFs, sometimes drawings. These are the most common casualty, because they are large, because they are fetched separately, and because a missing photo does not make the export fail. Open the media folder and count. If you recorded voice entries and there is no audio folder, stop and find out why before you go any further.

Everything the server computed. This is the pile people never think about. A map of where you have written, a year-in-review view, a mood or activity chart, an entry count, a streak, a "this day last year" screen, a shared or published link with its own URL, weather and place names attached to entries, a search index. Some of that is stored as fields on entries and comes along. Much of it is a rendering the server produced for you, and a rendering is not data. It leaves no trace in the bundle, and it stops existing when the account does.

The useful test is to ask, of anything you value in the app: is this something I wrote, or something the app worked out about what I wrote? The first almost always exports. The second usually does not.

That includes the links between entries and the outside world. If you shared an entry as a public page, that page lives on their servers, not in your archive. If your attachments are stored in a linked cloud drive rather than in the journal itself, then the export may contain pointers, and the actual files sit in an account you also have to keep. Follow one photo the whole way through to see which arrangement you have.

A note on where you land next

Before the counting, one question worth carrying into whatever you choose next. Ask it of any app before you commit years of writing to it: does export work on the free tier, in an open format, without asking anyone's permission?

That question is why Yesterself works the way it does. Export is free forever, including on the free tier, in formats other software can read, and there is no account and no email address to be cut off from. Yesterself is not released yet, so it is not somewhere to send an archive today. There is a waiting list on the site for when it is. The question matters more than the answer, and it is worth putting to the app you are actually choosing this week.

Verifying across the gap

Ordinary verification is counting. Verification for a synced app is counting twice, in two places.

1. Record the numbers from the server side while you still have the subscription: total entries, oldest entry date, number of notebooks or journals, and roughly how many attachments.
2. Count the same things in the bundle. Search the structured file for a field that appears once per entry and compare.
3. Test a cross-platform entry. Find something you know you wrote in the browser, or on the other phone, and confirm it is in the archive. This is the check that catches an incomplete mirror.
4. Open the far end of the rope. Your oldest entry. If the archive starts a year after you started writing, the export ran off a device that never held the beginning.
5. Open five attachments at random, from five different years, at full size. A folder with the right number of files can still be a folder of thumbnails.

Anything that fails here is repairable while you still have the account, the subscription, and the server copy. That is the entire reason to do it in this order.

What to keep before you stop paying

Beyond the entries themselves, a short list of things that exist only inside the running app and are cheap to capture now.

• Screenshots of the computed views you would miss: the stats page, the map, the year in review. They are pictures rather than data, but a picture is better than the nothing you will otherwise have.
• A written note of your structure. Notebook names, tag lists, what you used each one for. Rebuilding that from memory in a new app two years from now is worse than it sounds.
• Any shared or published links, copied as URLs and as saved pages, since the page goes when the account goes.
• The account details themselves. Which email address it was under, which sign-in method, which cloud drive was linked. This is the part that makes a re-export possible if the archive turns out to be short.
• A second copy of the bundle in a different place. A copy that lives only on the laptop you are about to replace is not a backup.

Then, and only then, in this order: verify, copy, keep the app installed and paid for one more cycle, cancel the subscription, and leave deleting the account for last. Deletion is usually the one step you cannot undo by paying again.

Afterward

Read the archive once, that first week, before you file it. Not to migrate anything and not to check anything. An export is the rare moment your writing sits in one flat pile, unsorted by any app's opinion of what mattered. Then put it somewhere boring and leave it there. A dull folder you can find beats a clever system you will not maintain.

Common questions

Does exporting from my phone get everything?

Not necessarily. In a cloud-synced journal the export is usually built from what is on that device, so a phone that has been reinstalled or is short on storage may hold thumbnails and recent entries rather than the full archive. Force a complete sync first, or export from the device with the most complete local copy.

What is in the exported bundle?

Typically the entry text plus a folder of attachments, in whichever format you chose. Things the server computed rather than stored as entries, like map views, statistics, streak counts and shared links, are usually not entry data and often do not appear at all.

Which format should I pick if I might switch apps later?

Take a structured export such as JSON or a zip archive, because it keeps dates, tags and the link between an entry and its photos. Take a plain text or Markdown copy as well so the archive stays readable without any software. Check what your version actually offers rather than assuming.

Do my entries disappear if I stop paying for sync?

That depends entirely on the app, and it is the thing to look up in writing before you cancel. In some apps the local copy stays and only syncing stops. Assume nothing, export and verify first, and leave the account deletion until last.

How do I check the export is complete before canceling?

Write down your entry count, oldest entry date and attachment count while you still have access, then check the archive against those numbers. Open entries from several different years, and confirm that anything created on another platform is present.