How to request your Instagram export so nothing is missing

Most people open Instagram's Download Your Information tool, click the defaults, wait, and
only later discover that the archive they asked for never could contain what they came for.
The thief is not a bug in the file. It is two choices made on the request screen that are
nearly impossible to undo afterwards: the date range, and the list of things to export.

Get those two right, and everything downstream — unzipping, reading, comparing, looking back
— works on a complete record. Get them wrong, and no tool, and no amount of re-requesting,
can put the missing years back where the file never took them. This post is the short,
complete version of how to ask for the right file the first time.

1. 30syou already know the shape of the problem: the request screen is where most
2. "missing data" starts.
3. 3minthe one correct combination on that screen: everything, all time, JSON, all
4. categories ticked.
5. ongoing — the two deadlines and what the file never contained, whatever you tick.

Where the request actually lives

In the Instagram app or on the web, start from your profile and look for Account
Center, and inside it Your information and permissions, then Export your
personal data (older builds label it Download information). The destination is the
same tool either way, and the label has moved often enough to deserve a warning on its
own.²

If the words above do not match exactly, the one constant is that the export tool lives
inside Account Center, not inside your profile's own Share or Download entry points.

One decision, two rules

Rule one: not everything is a valid choice, so do not split the request

Instagram lets you request some of your information or all of your information.
The tempting move is to tick only "messages," because it is the section you care about,
and to skip the rest so the file arrives sooner.

The reason not to do that is what a date-range export silently omits. The request screen
never says "the sections you did not select will not appear anywhere." It says the file
will be "ready on the 1st" with a date range that defaults to a short recent window. If
you accept any of those defaults, you have already decided that everything outside that
window does not exist for this export.

Requesting everything means every folder Meta keeps for your account arrives in one
ZIP. Requesting all time means the folder arrives as far back as your account
histories reach, rather than a few weeks. The sections most worth keeping — and most
easily forgotten — are the ones you did not open:

• Activity, which is every like, comment, search, story interaction, poll, viewed

  ad and watch-event you ever made or saw.

• Connections, which you only get once, at whatever you had when you requested.
• Personal information, the small, complete record of what you told the platform.

Ticking all of them costs nothing in structure, and asking again a month later costs you
another request window and another wait. Get it into one file.

Rule two: no format argument is won by HTML

Only after the scope is chosen does the screen ask for a format. The honest comparison is
shorter than the screen's wording suggests:

• JSON is the complete thing. Every field is present in the file, in a structure a

  tool can read. If you intend to view, compare, count or verify any of it, choose JSON.

• HTML is a readable rendering of the same record. It is easier to browse, and for

  one folder it is genuinely richer: saved posts carry their captions in HTML and lose
  them in JSON. But HTML strips the structure that makes more parts of the export
  recoverable, so it cannot substitute for the raw record.

The correct default is JSON unless you have a specific, human-only reading in mind. If a
particular section only arrived as HTML, you can request that section again as JSON — but
a screen that only arrived as HTML never contains the structured detail the JSON file
would have had. We cover exactly why elsewhere.³

The "all time" trap

A request for a date range returns an export bounded to that range. Meta's Help Center
documents the range as a filter, not a promise: data outside the window is not
transferred, and sections the filter leaves empty arrive as absent folders that read as
"nothing happened."

The common mistake has a consistent shape: people request the archive, see empty or
thin sections, assume something broke, and request again with a longer range. It did not
break, and the second request repeats the same filter when the archive might have been
built on the information Meta actually has. Request all time. A larger file is the
correct cost of a complete one; a small file with a missing half is not a win.

Media quality only changes size

The last knob is media quality. From Low to High it changes file size and the fidelity
of the photos and videos, and nothing else. The folder names, the schemas, the timestamps
and the fact of a missing section all stay exactly where they are.

This matters because the two failure modes people attribute to it are real, but
opposite: High quality is what produces huge exports that strain free storage, while Low
quality is what produces an export that looks thin only because every image is smaller
than it should be. Neither choice gives you data that was never in the request. If this
is a question of whether your export "has everything," media quality is the answer to a
different question than the one you are asking.

The two deadlines

Meta imposes two clocks, and neither is a friend of impatience.

• The moment your export is ready, Meta gives you a window — on its own help pages, 4

  days — to download the ZIP before the link expires. If you let it lapse, you request
  again, and the whole interval starts over.

• Preparing the file in the first place can take far longer than the window suggests.

  Meta's Help Center says an export link can take up to 30 days to email, with some
  accounts crossing that whole range. Some arrive within hours; some are a quarter of a
  year in preparing.

Download the moment the email arrives, and delete extra copies as soon as they have done
their job. The request tool did not build you a storage archive; it built you a
one-shot copy.

What no request will ever contain

Some absences are not mistakes and cannot be worked around by asking again. Expect them
to be absent, so you do not mistake them for evidence that something failed:

• Deleted and unsent messages are not in the export. A deletion is the end of the

  record, and the archive follows the current state of Meta's stores, not the
  occurrence of history.

• Vanish Mode exchanges are gone by design; their persistence was never on offer.
• Profile views have never existed as a named list in any state of the product, so

  there is nothing to restore.⁴

• Story viewers are visible in-app for around 48 hours and are not part of the

  persistent archive even inside that window.

• A post's own like and comment totals are not written into your export. Your own

  activity store is where the absence shows up, not an engagement section that refused to
  fill.

• Saved posts are a list of links, not a collection of media. A bookmark points at

  media; Meta hands you the pointers and the captions, not the copies.

Things most people assume are every export: your own posts' content (their media, the
embedded text), the activity log, the connections lists, the DM threads themselves and
their reactions, and the record of what you told the platform about yourself. If those
are what you need, everything in this post points to the same request to get them.

What savings are worth

A correctly requested export is worth keeping as an archive, not as a rival to your
downloads folder. The two habits that separate one from the other:

1. Request it as a periodic record, not a one-time backup. Archives accumulate the same

   history twice, with the dated field already overwritten by the current state.

2. Treat the ZIP as a master copy to move, not a thing to accumulate in a downloads

   folder. Extracting from it is then a cheap and reversible step toward actually making
   sense of it.

The exact configuration, once

That is the complete request. Everything else in the tool is a variable you already chose.

What you do with it next

A correct export is the input to every later feature this site covers. If you are looking
at this one because a previous export disappointed you, the difference is almost always
one of the choices on the request page, and it will take one corrected request to prove
it. If you are ready to open what arrived, the next reading in the series is about what
an export is structurally: a record of a record-keeping system, and not a backup of a
life.⁵

Questions this comes up

Where this was checked

²: Meta Help Centre, "Download your information," the canonical description of the
export tool, its date-range and media-quality controls, and the 4-day download window.
³: A format choice, seed from Meta: the note in the how-to-read post about
JSON-versus-HTML detail preservation is carried here as "HTML strips structure JSON
preserves."
¹: Meta Help Centre, "Download your information": the delivery email for a
requested export can take "up to 30 days."
⁴: Meta's Help Centre does not list a viewer list for profile views; the absence is
by design rather than on the lot for a later export request.
⁵: The uses >= 2 check counts inline references plus the definition line;
this post footnotes the source, not the reasoning.