Does Premium actually boost reach?
Not as a direct multiplier in the released ranking code. There is no "is_verified" or "is_premium" term in the scorer that multiplies a subscriber's reach. What the code does have is subscription as a graph feature (who you subscribe to) and a filter for subscriber-only content — not a reach bonus. Any reach difference Premium users see routes through indirect effects (more posting, longer posts, different engagement), not a verified-boost line in the formula. The honest answer: the open code doesn't contain the boost the marketing implies.
"Get verified, the algorithm boosts you" is widely believed and widely sold. As a claim about the released ranking code, it doesn't hold up the way people assume — and the precise version of the truth is more useful than either the hype or the backlash.
No verified multiplier in the scorer
We searched the weighted scorer and the candidate features for an is_verified,
is_premium, or subscription-tier term that multiplies reach. It isn't there. The scorer
sums predicted engagement actions; none of them is "author has Premium."
The released ranking pipeline contains no link-penalty, URL-deboost, or external-link downranking mechanism. The only link-related signal is click_score in the weighted scorer, which is a positive term. Claim is scoped to the open release at the pinned commit.
What subscription actually is in the code
Subscription appears in two places, neither a reach bonus:
| where it appears | what it actually does |
|---|---|
subscribed_user_ids in viewer features | A graph signal: accounts you subscribe to. It personalizes your feed toward them — it doesn't boost their global reach. |
| Ineligible-subscription filter | Gates subscriber-only content from people who aren't subscribers — access control, not promotion. |
The home mixer includes a family of query hydrators that load viewer context before ranking: followed/blocked/muted/subscribed user IDs, muted keywords, impressed and served history, mutual-follow overlap, IP/country, and (gated) demographics and inferred gender.
Why Premium users still sometimes see more reach
The effect people observe is real but indirect — and routes entirely through signals already in this library:
| indirect path | the real mechanism |
|---|---|
| Longer posts & more features | Premium unlocks longer posts and editing, which can earn more dwell — and dwell is a rewarded signal. |
| Reply prominence in their own threads | Subscriber reply features can surface their replies — engagement, not a ranking multiplier. |
| More posting | Engaged subscribers post more, creating more chances to be retrieved. |
What the code doesn't say
Whether a verified boost exists outside the released ranking pipeline. Our claim is
bounded: no direct multiplier is in the open scorer at the pinned commit. X could weight verified
status in a layer that isn't published, and the withheld parameters could in principle carry such a
term. We can say it's not in the open formula; we can't prove it exists nowhere.
The numeric values of the current weights are not included in the open-source release: weighted_scorer.rs references a params module (e.g. p::FAVORITE_WEIGHT, p::REPLY_WEIGHT) whose values are not present anywhere in the published repository.
What to do with this
Decide on Premium for its features (longer posts, editing, reply tools), not for a ranking boost the code doesn't show. If the features help you earn dwell and replies, that's real — but it's the engagement doing the work, which means the same content discipline matters with or without the checkmark. xDoctor measures the signals that actually move ranking, regardless of subscription tier.