Your Mutual Followers, From Your Own Export

There is no sampling and no API limit here. Whether you follow two hundred accounts or two hundred thousand, the overlap is computed over every row, not over a page of results, and it is free at either size. What it cannot outrun is a truncated archive. An export limited to a date range arrives with its followers list already filtered, and the mutuals missing from it look exactly like people who never followed you back.

If you are trying to find the mutual followers between two other accounts — yours and a friend's, or two public profiles — this is the wrong tool. That needs read access to accounts that are not yours, which means either scraping Instagram or holding your login. This tool does neither, ever; the code is open source (MIT), so you can confirm that rather than take our word for it.

What it does instead is answer the version of the question you have the data for: your own, exactly, from the export you already hold.

Both are in the connections/followers_and_following folder of the archive Meta builds for you: following.json and followers_1.json, or their .html twins. Where that folder sits in older archives, and every step of requesting one, is on the export guide. Choose All time rather than a date range when Instagram offers you the choice — a truncated list on either side removes mutuals that are really there.

Mutuals as a share of the people you follow is the number most people are actually after. A low share is not a verdict on anything: following many accounts you do not know — brands, artists, news — drives it down by design.

The comparison is only meaningful against your own earlier archive, and only if both were exported the same way.