Your Instagram data download is not the sprawling archive people expect. If you requested just Followers and Following (as the export guide recommends), the ZIP holds a handful of small JSON files, usually well under 1MB combined. Here is exactly what is in it.
The core files
- followers_1.json (or followers_2.json, followers_3.json for larger accounts): every account that follows you
- following.json: every account you follow
- pending_follow_requests.json: accounts you have asked to follow who have not accepted yet, if any
- recently_unfollowed_profiles.json: a short, Instagram-maintained list of accounts you recently unfollowed
That last one is worth pausing on. Instagram keeps a small rolling log of your own recent unfollows, but it is short and temporary, not a substitute for tracking who unfollowed you over time yourself.
What each entry actually looks like
Every account in these files is a small JSON object, not just a username. Each one carries the same three fields:
- value: the username
- href: a direct link to their profile, like https://www.instagram.com/username
- timestamp: a Unix time value marking when the follow relationship was recorded
Three fields, every time: the username, a direct link to their profile, and a timestamp.
Why the timestamp is the important part
The username and link are easy to guess. The timestamp is not, and it is the one field that actually unlocks anything. It is what lets a tool calculate how long someone has followed you, tell a brand-new follower from a two-year one, and build a growth timeline instead of a flat list. This only shows up reliably if you chose JSON format. HTML exports carry the same relationships but usually drop the timestamp entirely, which is exactly why JSON is the recommended format.
What is not in there
No password, no private messages, no photos, nothing beyond the connections themselves if you selected only Followers and Following. Instagram's full archive (which nobody needs for this) can include posts, stories, and message history, several gigabytes of data you do not have to touch. The focused export is small on purpose.
Why this matters for a privacy-first tool
Because the whole file is just a few small, plain JSON documents, a tool like WhoUnfollowed can read it entirely inside your browser in about two seconds. There is no server involved because there does not need to be. The parser that reads these exact files is open source (MPL-2.0), so anyone can check that this description of the file matches the code that touches it.
