How we rank what you see

• Discovery is the default mode in your feed. Posts are ranked by an algorithm that combines how engaging the post is with how fresh it is and what you've already seen. The next bullets break that down.
• Engagement rate, not raw counts. A post saved by 3 of 10 viewers (30%) ranks above one saved by 50 of 5,000 viewers (1%). This deliberately lets a small-account post compete with an established-account post: the math normalises for exposure.
• Engagement signals are weighted by what they say about quality. Saves count 4×, comments 2×, shares 1.5×, likes 1×. A save (you want to come back to a post) is the strongest signal of quality; a like is the easiest, weakest one.
• A new post starts with a visibility floor that fades by ~50 views — without this, brand-new posts (zero engagement, zero views) would never surface. An older post decays in the opposite direction: its score halves roughly every two days, so the feed keeps cycling.
• The order shifts on every refresh. A small random jitter (about ±15%) reshuffles posts with similar scores, and posts you've already scrolled past get pushed down so new content can take their place. Both happen per-refresh, per-user.
• Following doesn't use any algorithm. Posts from accounts you follow appear newest-first, full stop. No quality filter, no hidden ordering, no suggested posts mixed in. The day Following starts re-ordering posts behind your back is the day it stops being yours.
• Ranks (the trophy at the top of the feed) is a separate monthly competition. Posts compete inside the calendar month they were posted in; once the month closes the rankings are frozen. Tap How does Quality Score work? inside Ranks for the visual breakdown of the formula and the vision behind the future prize pool.

Frequently asked questions

What's *not* in the Discovery score?

Follower count, account age, posting time, posting frequency, and anything paid. The score only looks at how the post performs relative to its views, plus the per-refresh freshness and seen-history layer. None of those excluded factors affect what surfaces.

Why does the order change between refreshes?

Three things. Older posts lose score gradually (time decay). A small random jitter shuffles posts with similar scores. And posts you've already scrolled past get pushed down so new content can surface. Combined, they keep the feed from feeling stuck.

Does a brand-new post stand a chance against established work?

Yes. New posts get a guaranteed visibility floor that fades as views accumulate. After that they have to compete on engagement rate — which normalises for exposure, so a 100-follower artist's standout post can outrank a 10,000-follower artist's average one.

Do my own likes, saves, and comments on my post count toward the score?

No. The displayed counts include them (so the number you see matches what other viewers see), but the algorithm subtracts them before computing the score.

Can someone game it by buying views or repeat-viewing?

Repeat views from the same account are capped (a single user can't inflate a post's view count beyond a small daily limit). Saves — the heaviest weight — are also much harder to fake than likes. Beyond that, the formula is published: there's no hidden lever to pull. The only signal that ranks high is real engagement.

Why is Following so simple compared to Discovery?

That's the point. When you choose to follow someone, you've already done the curation. Showing their posts in order respects that decision. Mixing in algorithmic ordering or suggested content would override it.

Will the algorithm change?

When it does, we'll publish what changed. The in-app Quality Score page already includes a version-history section. We'd rather argue about a documented change than have you guess what shifted.