Tracking Instagram Stories Before They Disappear

A post lives on a profile indefinitely — pull it today, pull it again next month, same data. A Story doesn't work that way. It's live for roughly 24 hours, and once it expires there's no public record of it left to fetch. If a competitor posts a discount code, a flash-sale link, or a product teaser to Stories only, the only window to capture it is while it's actually live.

That's a different data problem than tracking posts. It's not about how much history you can pull — it's about how often you check, and whether you check before the thing you're watching disappears.

If you're an agent (or building one) reading this rather than a human, the full machine-readable schema for every endpoint below lives at mindcase.co/skills.md.

Why does Stories data need different handling than posts?

Posts, followers, and profile data are all durable — pull them whenever, the record's still there tomorrow. Stories are the one Instagram surface where "I'll check it later" is a real failure mode, not just a delay. A tool built to poll posts once a day, on the same schedule, will miss every Story that goes up and comes down inside that gap.

The practical effect: if you care about Stories at all — a competitor's promo drops, an influencer's sponsored placements, a brand's behind-the- scenes content — the polling interval isn't a performance detail, it's the feature. Check every few hours and you catch most of what goes up. Check once a day and you're relying on luck.

What does the Instagram Stories API actually return?

Instagram Stories API takes one or more usernames and returns each account's currently-active Stories — nothing more, nothing retroactive. An account with no live Story returns no rows for that pull; there's no way to ask for what someone posted yesterday.

Each row is 21 fields: storyId, shortcode, username, profileUrl, fullName, isVerified, mediaType, imageUrl, videoUrl, videoDuration, hasAudio, width, height, takenAt, expiringAt, isPaidPartnership, linkUrl, linkTitle, musicTitle, musicArtist, posterProfilePic. Two of those are the ones that matter most for scheduling: takenAt tells you when it actually went up, and expiringAt tells you exactly when it's gone — not an assumed 24 hours, but the real timestamp Instagram is enforcing for that specific Story.

How do you pull an account's active Stories?

Say you're tracking a competitor's promotional activity and they've started running flash sales through Stories instead of the main feed.

curl -X POST "https://api.mindcase.co/v1/data/instagram/stories/run?wait=true" \
-H "Authorization: Bearer $MINDCASE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "params": {
    "usernames": ["competitorbrand"]
  }
}'

# $0.01 per story

usernames accepts an array, so one call covers a whole watchlist — a handful of competitors, or every account in a sponsorship deal you're auditing — not one account per request. linkUrl and linkTitle are where a Story's swipe-up/link-sticker destination shows up, which is usually the actual payload you're after in a promo-tracking pull: the discount code or landing page a competitor only ever put in front of their Story audience, not the main feed.

How often should you poll, and why does it matter?

There's no webhook or push notification for a new Story — the only way to know one exists is to ask. That makes the polling interval a direct trade-off against $0.01 per story returned: poll every hour and you'll catch nearly everything but pay for a lot of empty checks against accounts that haven't posted; poll once a day and you'll miss anything that went up and expired inside that window.

A reasonable middle ground for competitor or influencer monitoring is checking every 2-4 hours during the hours an account is actually active — takenAt on what you do catch tells you their real posting pattern, so the schedule can tighten around it after the first few pulls instead of guessing upfront. expiringAt is also worth storing even when you can't capture the media itself — a log of when an account posted Stories, and how many, is its own signal about posting cadence and campaign timing.

What can't you get from Mindcase?

Two cases.

You don't already know the account. usernames is the only way in — there's no keyword or hashtag search across everyone currently posting Stories about a topic. You need the specific handle before you can check it, same as you'd need to open the app and know whose Story you're tapping.

A Story that already expired. Once expiringAt passes, that Story is gone from Instagram's own servers, not just from this endpoint — no vendor can return media Instagram itself no longer serves. If a pull runs late, the miss is permanent for that Story; the fix is a tighter polling schedule going forward, not a longer lookback window.

Which endpoint should you use for which job?

FAQ