Does X Shadowban New Accounts? What X Publishes (2026)

Twitter (X) does not publish any rule that shadowbans a new account because it is new. Neither X's Help Center nor its open-source code, both read on November 5, 2026, sets a probation period for new accounts. They do contain a boost for small authors, filters aimed at new accounts and spam rules for all accounts.

Many people report that a new account gets no views for days. This guide separates what X publishes from the rumors, and shows how to start an account without looking like spam. X gives no number of days for a new account, so this guide gives none either.

Does X shadowban new accounts?

No published rule of X shadowbans an account for being new. X's published code has no reach penalty tied to the age of the author, and X's Help Center describes no trial period for new accounts. X does not publish the learned weights of its ranking model, so the code cannot rule out an effect the model learned.

Three things do exist, and they are often mixed up with a shadowban:

• A boost for posts from small authors, whatever the age of the account.
• Filters that other users can turn on to hide new accounts from their notifications.
• Spam and automation rules that apply to every account, and that new accounts often break by accident.

The code shows defaults, not production. X's home ranking file says its values are "mirrored from config feature-switch defaults; last sync 2026-11-04T16:00:31Z". X can change them at any time.

Does X have a waiting period for new accounts?

X publishes no waiting period for the reach of new accounts. The one it mentions is on its X Premium page. The Availability section says, after a sentence on new Premium subscriptions: "We may also impose waiting periods for new accounts in the future at our discretion, and without notice."

That sentence is about Premium, it speaks of the future, and it gives no length.

X can also lock an account while it is being set up. Its page on locked and limited accounts says: "If you're setting up a new account and get locked out, contact our support team." That is a lock, not a reach restriction.

Does X give new accounts a boost?

X gives a boost to small authors, and the age of the account is not one of its conditions. X's README calls it the New-Author Boost: posts from authors with few impressions are "lifted toward a target position".

With the defaults published in X's code, a post qualifies only if all of these are true:

The step is on by default. The code picks one post per feed request among those that qualify, and raises its score to at least the score of the 16th post. It never lowers a score.

This is not a promise of views. For viewers in X's test groups, only authors of the matching group qualify, and one post per request is a small share of a feed. Our guide on what X's code says about reach explains the other reach rules. What X itself says about limiting reach is in does X shadowban accounts.

Which filters in X's code treat new accounts differently?

Four rules in X's code concern new or small accounts. Two of them concern the reader, not the author, which is easy to miss:

X's README describes the second rule as: "For new accounts, out-of-network posts below an engagement threshold". In the code, the age checked is the age of the account that asks for the feed, or of one that has just returned. It shapes what a brand-new account reads. It does not limit what a new author reaches.

The third rule removes out-of-network posts labeled SOFT_NSFW from For You, for readers under 7 days old with a short history. It is also about the reader, and off by default.

Why do replies and mentions from a new account not notify people?

Because the person you mention can turn on notification filters against accounts that look new or incomplete. X's page on reach limits says people can filter their notifications "from certain types of accounts, like those without a profile photo, unverified email address or phone numbers, and from accounts they do not follow."

X's code names five such filters, read from the recipient's own settings:

• no confirmed email;
• no confirmed phone number;
• default profile image;
• new users, which means an author whose account is younger than 72 hours;
• authors who do not follow the recipient.

A sixth setting limits notifications to people the recipient follows. The filters apply to mentions, replies, tagged media and new follows, and only when the recipient does not follow you. A recipient who filters default photos is not notified of your mention if you have no photo.

The notification rules also drop authors with certain account labels, with no setting needed, for recipients who do not follow them. One is named NotGraduated. The published files do not describe it, so we do not guess what it means.

These are notification rules, not search or timeline rules.

How long is a new X account restricted?

X does not say how long a new account is restricted, because it publishes no restriction for new accounts. Its page on locked and limited accounts says a limit lasts "for a specified time, which varies based on the nature of the violation".

The closest numbers in X's code concern spam labels. In the enforcement config mirrored on October 20, 2026, several rules add SpamHighRecall to an account for 7 days or for 30 days, and one rule, for a spam campaign, sets no lifetime. X's label text says: "The account's posts are hidden from recommendations to non-followers." It applies to accounts of any age.

A lifetime in code is not a recovery date. X can apply a label again if the same behavior is detected. Any fixed number of days quoted for a new account does not come from X. Our guide on how long a shadowban lasts covers the durations we measure.

What makes a new account look like spam to X?

Behaviors that X's rules call spam or inauthentic activity, which apply to every account. New accounts often trigger them while trying to grow fast.

A limited state means "your account and posts may be filtered out of certain places on X, including from search results and notifications". The technical follow limit is 400 per day, and posting is capped at "50 original posts and 200 replies per day for unverified accounts".

When X detects suspicious levels of activity, accounts "may be locked and prompted to provide additional information (e.g., a phone number) or to solve a puzzle".

X's code also has an account label for spam, described as "Account detected by automated systems as likely to post spam."

How do you warm up a new X account?

X does not describe warming up an account. It publishes behaviors to avoid and settings that other people's filters read, so the steps below follow those rules. They are not a tested recipe.

Guides in Chinese call this 养号 (raising an account), and Japanese guides talk about 新垢 (a new account).

• Add a real profile photo. A default photo is one of the five notification filters. X's Authenticity policy also names "stock, stolen or AI-generated profile photos" among fake personas, when used "for the purpose of deceiving others".
• Confirm your email and phone number. Two notification filters read them. X also says: "Keep this information updated. You'll need it to restore access if your account becomes locked or limited."
• Write your own bio. The same policy names "copied or stolen profile bios" among fake personas.
• Post original posts first. The New-Author Boost applies to original posts only, not to replies or reposts.
• Follow people gradually, in your topic. This avoids "indiscriminate following" and the 400 per day limit.
• Reply where you add something. Avoid the same reply, the same link or a long list of mentions.
• Use few hashtags. X recommends 2 at most.
• Use only approved apps. X's automation rules allow automation through the API and forbid scripting the website.

Do not open a second account to escape a restriction. X's Authenticity policy lists "creating new accounts" as a way to evade enforcement.

Why does a new account get no views or not show in search?

Often for reasons that are not a restriction. X lists some. Its page on reach limits describes posts that do not appear and replies that detach. It says these cases "may be the result of a post being identified as spam by our systems". It adds that they may come from "current technical issues and limitations of the service".

A new account also has few followers, so most views come from recommendations, where the rules above apply. Our guide on posts missing from X search lists the causes, and the one on X's temporary label and limits explains the messages X shows.

How do I check an X account less than 1 year old for a shadowban?

Run Shadowban Radar's free check: it works on any public account, whatever its age. X's Under the Hood only covers accounts at least 1 year old with at least 10 posts that month, so it shows nothing for a newer account.

The free check needs no login. It tests the 4 restrictions people call a shadowban: search suggestion ban, search ban, ghost ban and reply deboosting. It shows what other people see today. It does not say why. Our page on X's Under the Hood explains what X's own report covers.

Sources

Read on November 5, 2026, on X's Help Center and in the xai-org/x-algorithm repository at commit 021afa4, published November 5, 2026.

• X Help Center, Authenticity (April 2025): spam, follow churn, hashtags, fake personas, multiple accounts, ban evasion and anti-spam challenges.
• X Help Center, About X limits: daily limits on posts, replies, messages and follows.
• X Help Center, locked and limited accounts: locks at account setup, limits for a specified time and what a limited state filters.
• X Help Center, specific instances when a post's reach may be limited: notification filters and posts that do not appear.
• X Help Center, automation rules: scripting the website and automated activity.
• X Help Center, how to use hashtags: the recommendation of 2 hashtags per post.
• X Help Center, About X Premium: the sentence on waiting periods for new accounts.
• author_cold_start.rs: which posts and authors qualify for the boost.
• author_cold_start.rs: post age, views and rank conditions.
• param.rs: the boost defaults, with the "last sync" header on line 1.
• param.rs: the new-account filters and their off-by-default switches.
• new_user_min_engagement_filter.rs: the filter reads the age of the requesting account.
• new_user_soft_nsfw_filter.rs: the reader-side filter for readers under 7 days.
• README.md: the description of the engagement filter.
• levels.rs: the five notification filters and the account labels dropped in notifications.
• author_rules.rs: the notification filters, active only when the recipient turns them on.
• author.rs: an author is new for 72 hours.
• enforcement_user.yaml: spam labels set for 7 or 30 days.
• underTheHoodLabels.strato: the effect of the account spam label.

Shadowban Radar is not affiliated with X Corp.