How to fix a shadowban on X (Twitter)

We have all experienced this at least once: overnight, an X account's impressions collapse. Likes vanish, replies generate nothing, tweets that would normally get 5,000 or 10,000 views plateau at 200. The account is still active, posts go through normally, nothing seems broken. And yet, nobody sees you any more. What X will never tell you is that your account has probably been shadowbanned.

X does not use this term in its official documentation. The platform talks about "visibility filtering", "reach limits" or "search result restrictions". But the outcome is the same: your content is made invisible or nearly invisible to non-followers, with no notification, no warning and no explanation. After helping several accounts recover and watching the system operate for months, we can say that shadowbanning on X in 2026 is more frequent, more subtle and easier to trigger than most people think.

The 4 types of shadowban on X

What most people do not know is that there is not a single shadowban but four distinct restrictions, each affecting a different surface of your visibility. You can be hit by just one, or by several at the same time.

The Search Suggestion Ban is the mildest form. Your handle disappears from the autocomplete dropdown when someone starts typing your name in the search bar. People have to type your exact username to find you. It is the most common restriction in 2026 and also the one most users never notice, because it does not directly affect your tweets. It typically lifts within a few days.

The Search Ban is far more serious. Your tweets no longer appear in search results, not by keywords, not by hashtags, not even when someone searches for the exact text of your post. This is the restriction people mean when they say "I am shadowbanned". On the ground, we observe that it is almost always triggered by behaviour X interprets as spam: repetitive links, identical replies, hashtags unrelated to the content.

The Ghost Ban (or Thread Deboost) is the most insidious. Your replies in conversations are hidden behind the "Show more replies" button that nobody ever clicks. You reply to a popular thread, you think your reply is visible, but in reality it is buried. According to X's own DSA transparency report, accounts under visibility filtering experience an 82 to 85% drop in impressions. That is colossal.

Reply Deboosting (or Quality Filter Discount) systematically pushes your replies to the very bottom of threads, under replies from accounts deemed more "trustworthy" by the algorithm. You are not hidden, but you are pushed so far down that nobody sees you unless they scroll extensively.

How to tell if you are shadowbanned

The most reliable test requires no tool at all. Log out of your X account (or open a browser in private mode), then type from:yourusername in the search bar and click the "Latest" tab. If your recent tweets do not appear even though they are published on your profile, you have a Search Ban. Next, start typing your handle in the search bar without pressing enter: if your name does not appear in the suggestions, you have a Search Suggestion Ban.

For the Ghost Ban, it is trickier. Ask someone who does not follow you to open a thread you replied to. If your reply is hidden behind "Show more replies" or simply missing, you are ghost banned.

There are also online tools that automate these checks. What the technical specs of these tools do not tell you is that none is 100% reliable because X does not publish any official shadowban status. Tools based on X's official API give more consistent results than scrapers, but even they can miss certain restrictions, especially Reply Deboosting which is very difficult to detect automatically.

An indicator that is often more telling than any tool: look at your impression statistics directly in X Analytics (or in the "View post analytics" tab under each tweet). If your recent impressions are near zero while your content and posting frequency have not changed, and your older tweets still show normal figures, that is active suppression in progress.

What triggers a shadowban in 2026

After observing dozens of cases, we find that triggers cluster into a few recurring categories. And what is frustrating is that some perfectly normal human behaviours are interpreted as spam by the algorithm.

Repetitive external links are the number one trigger in 2026. Posting the same link multiple times, even if you vary the surrounding text, is a major red flag. What many people do not know: putting a link in the first reply to your own tweet rather than in the tweet itself is a technique the algorithm now recognises. If you do it on every single tweet systematically, X spots it and treats it as a spam pattern.

Identical or similar replies sent to multiple accounts are another classic. Replying "Great article!" or "I agree" to 30 people in an hour is exactly the kind of behaviour bots reproduce. X does not distinguish the intention behind the message; it detects the pattern.

Mass follow/unfollow remains a toxic signal. Following 200 people in a day then unfollowing 150 the next is the textbook pattern of "growth hacking" tools and X catches it immediately.

Third-party automation tools are a common trap. Even tools that market themselves as "safe" or "approved" can trigger a shadowban. In March 2026, X ran a mass suspension wave for "inauthentic behaviour" caused by a new spam filter that misfired for about 12 hours. Perfectly legitimate accounts were caught in the net simply because they used analytics or tweet scheduling tools connected via overly broad permissions.

A consistently aggressive or combative tone is a signal many people overlook. Even with high engagement, systematically aggressive or controversial content leads to algorithmic deboosting. Negative engagement (people clicking to reply in anger) is not worth the same as positive engagement in the eyes of Phoenix, the transformer-based ranking model that replaced the old HeavyRanker in January 2026.

Abusing trending hashtags unrelated to your content is another classic trigger. Slapping #BreakingNews or #Trending on a tweet that has nothing to do with current events is exactly what spammers do, and the algorithm makes no distinction of intent.

The protocol to recover

Let us be direct: there is no magic button to lift a shadowban. No dedicated complaint form, no specific support contact. The only method that works consistently is to remove what triggered the restriction and wait for your account's trust score to climb back up. Here is the protocol, in order.

Step 1: stop all activity immediately. No tweets, no likes, no replies, no retweets for a minimum of 48 to 72 hours. This is not abandoning your account; it is a signal sent to the algorithm that the problematic behaviour has stopped. We know it is frustrating, especially if you are convinced you did nothing wrong. But on the ground, this initial pause is the essential first step.

Step 2: clean up your account. During this pause, delete the content that may have triggered the issue. Tweets with links posted multiple times, serial replies, tweets with unrelated hashtags, anything that could be interpreted as spam or that violates X's terms of use. Do not panic-delete your entire history; target only recent problematic content.

Step 3: revoke third-party apps. Go to Settings, Security and account access, Apps and sessions, Connected apps. Remove every app you do not recognise or do not need. "Growth" tools, unofficial analytics tools, automation tools, browser extensions for X: all of it has to go. Even if you think they are harmless, every connected app with broad permissions is a risk signal for the algorithm.

Step 4: check your official status. Go to Settings, Your account, Account status. If X has applied a visible restriction (a label), you will see a message with a "Request a review" option. This option requires a Premium subscription (formerly Twitter Blue). If no label appears but your impressions are still on the floor, it is an algorithmic shadowban without a label, the most common type.

Step 5: resume activity gradually. After 48 to 72 hours of pause, start publishing again, but slowly. One tweet per day maximum during the first week, no external links, no hashtags, no serial replies. Write original, engaging content that generates positive interactions (likes, retweets, constructive replies). The goal is to rebuild your account's trust score in the eyes of the algorithm.

In most cases, the mildest restrictions (Search Suggestion Ban) lift within 24 to 72 hours. Search Bans typically take 3 to 7 days. Ghost Bans can persist for 7 to 14 days. For persistent cases beyond 14 days, some users report that a complete logout for a month (cookies cleared, zero activity) eventually resets the score, but that is the last resort.

What does not work (and is recommended everywhere)

Avoid the following advice; it circulates everywhere but changes absolutely nothing. Reinstalling the X app does not reset your status because the shadowban is tied to the account, not the app. Sending a DM to X support does not trigger any specific action for algorithmic shadowbans. Creating a new account to "start fresh" is often worse: X detects accounts created from the same device or IP and can apply even harsher restrictions to the new account.

You also see plenty of recommendations like "tweet positive content" or "be authentic". Well-intentioned but useless in the short term. The algorithm does not measure your authenticity; it measures behavioural patterns: posting frequency, link types, engagement-to-impression ratio, quality of interactions received. It is mechanical, not subjective.

How to avoid a shadowban in the future

After watching accounts fall and recover, we can summarise best practices in a few simple rules. Space out your interactions: no more than 20 to 30 replies per day, no bursts of likes, no mass follow/unfollow. Vary your content: do not post the same link or tweet format systematically. Limit external links: if you need to share a link, surround it with substantial original content, and do not share the same link more than once. Remove connected apps you are not actively using. And if you feel your impressions dropping unusually, do not wait: slow down immediately for 24 hours. That small preventive pause can save you an entire week of suppression.

A shadowban on X is not a permanent punishment; it is an algorithmic signal saying "this account is behaving suspiciously". In the vast majority of cases, the restriction lifts on its own within a few days if you stop the behaviour that triggered it. The protocol is simple: a 48 to 72-hour pause, cleaning up problematic content and apps, gradual return with original content. What does not work: panic, over-correction, creating a new account. What works: patience, consistency, and removing anything that remotely resembles automated behaviour.