The One Format That Dominates Everything Else
If you want more impressions on X, there is one format that is not even close to the rest.
Talking videos - original video with a real person speaking directly to camera - averaged 1,715,897 views per post in our analysis of over 3,400 tweets. The next best format was not within shouting distance.
List and bullet posts averaged 27,643 views. Number-hook openers averaged 6,723 views. Threads averaged 1,353 views. Personal story posts came in at 3,641 views.
Talking video gets roughly 62x more views than the average thread. A category difference.
This was validated at the platform level. X product head @nikitabier posted: "People should make more talking videos like this on X. Easy way to get millions of impressions with basically no followers." That post earned 11,393 likes and 13.5 million views - the single highest-performing post in the dataset.
X is pushing talking video right now - a real person, real voice, direct to camera. Produced B-roll, stock clips with text overlays, and repurposed content from other platforms are not the same thing. The algorithm treats them differently.
The 90% Penalty Creators Are Getting Hit With
Here is the most important algorithmic fact you need to understand right now.
If your post is a repost or sourced from a third-party network, X applies up to a 90% deduction on impressions. It was stated directly by @nikitabier in a post that earned 4,708 likes and 963,000 views.
A separately viral post corroborated this with a tiered breakdown: aggregator accounts receive a 60% revenue reduction. Reposts receive a 90% impression reduction. Accounts that habitually post "BREAKING" bait content receive permanent deductions.
If you are reposting viral clips from other platforms, screenshotting posts from Instagram or TikTok, or sharing content you did not create, you are handing X a reason to suppress every post on your account. Not just that post. Every post.
Growth advice you will find still recommends curating and sharing good content. That playbook is now actively penalized. X changed the distribution rules to reward original content only. Original voice, original video - and your perspective needs to be yours, not a repackaged take from somewhere else.
Tweet Length Data
I see this everywhere - advice telling you to write longer, more detailed posts. The data says something different.
Short tweets under 140 characters averaged 36,257 views. Medium tweets between 140 and 280 characters averaged 8,881 views. Long tweets over 280 characters averaged 14,343 views.
| Length | Avg Views | Avg Likes |
|---|---|---|
| Short (under 140 chars) | 36,257 | 93 |
| Medium (140-280 chars) | 8,881 | 91 |
| Long (280+ chars) | 14,343 | 61 |
Short tweets outperform medium tweets by more than 4x on views. They outperform long tweets by more than 2.5x. The pattern here is clear: punchy and short works, or long and substantive works. The middle ground - the paragraph-length post most people default to - performs worst of all.
If you cannot say it in under 140 characters, go long. Do not stop in the middle.
Content Format Breakdown
Here is the full view of how different content formats perform:
| Format | Posts Analyzed | Avg Likes | Avg Views |
|---|---|---|---|
| Talking video | 9 | 2,347 | 1,715,897 |
| List/bullet posts | 87 | 116 | 27,643 |
| Number-hook openers | 85 | 150 | 6,723 |
| Personal story | 75 | 95 | 3,641 |
| Questions | 182 | 52 | 2,340 |
| Threads | 64 | 39 | 1,353 |
The data is not saying threads and personal stories are useless. They are fine for building a presence. But if you want raw impressions, nothing competes with original video with your voice on it.
The Reply Guy Strategy Works - But It Has a Hard Ceiling
People are documenting it and the numbers hold up.
Find Your Next Customers
Search millions of B2B contacts by title, industry, and location. Export to CSV in one click.
Try ScraperCity FreeIn one tracked case, an account running 20 or more replies per day for 28 days reported gaining 19,000 followers, 12 million impressions, and an extra $2,000 in revenue. Another documented experiment running 50 or more replies per day for 14 days claimed 102 million impressions.
In a smaller-scale test, one account with 757 followers sent 315 or more replies in a single day and logged 13,000 impressions plus 37 new followers from that effort. Day two of a tracked experiment with a roughly 6,000-follower account showed 69,936 impressions in 48 hours and 239 new followers.
So yes - replying heavily to large accounts works. It gets your name in front of new audiences. It drives profile visits and follower growth.
But here is the ceiling: reply-guy tweets average only 66 likes and 2,232 views in our data. The dataset mean is 148 likes. Replies underperform standalone posts by a wide margin on their own metrics.
The reply strategy generates impressions on other people's content. You appear in their comment sections, you pick up followers, but you are not creating standalone viral posts. You are renting attention, not owning it.
The accounts that grow the fastest are doing both. They reply strategically to large accounts to build name recognition, and they post original talking videos to build a post catalog that earns impressions on its own.
The Engagement Rate Paradox at Large Follower Counts
Here is something that surprises almost everyone who sees it for the first time.
Accounts with 200,000 or more followers have an engagement rate of approximately 0.22%. Accounts with 10,000 to 50,000 followers have an engagement rate of roughly 5.51%. Accounts in the 50,000 to 200,000 range hold at about 5.25%.
| Follower Tier | Avg Likes | Avg Views | Engagement Rate |
|---|---|---|---|
| Nano (under 1K) | 6 | 295 | 3.05% |
| Micro (1K-10K) | 32 | 1,535 | 2.93% |
| Small (10K-50K) | 74 | 2,595 | 5.51% |
| Mid (50K-200K) | 160 | 5,224 | 5.25% |
| Large (200K+) | 343 | 204,259 | 0.22% |
Large accounts get massive raw view numbers, but almost nobody in their audience interacts. The algorithm notices this. A low engagement rate tells the algorithm your audience is not reacting to your content - and it throttles distribution accordingly.
Chasing follower count can hurt your impressions per post if your engagement rate collapses. The 10K to 200K range is where accounts punch hardest relative to their size. Growing too fast by gaming follows without building real engagement can move you backward.
The practical takeaway: growing to 50,000 genuine followers who engage with you will outperform 200,000 passive ones on almost every metric that matters to the algorithm.
The 5 Million Impression Threshold Is Changing How People Post
X's creator monetization program requires hitting 5 million impressions to unlock revenue sharing. Posting behavior across the platform is shifting around this number.
In our analysis, 123 posts directly referenced the "5M impressions" threshold. Creators are explicitly building their posting strategy around this number. That means posting 20 to 25 times per day, engineering comment section engagement, running coordinated reply threads, and documenting their progress publicly.
There is a secondary opportunity here. When someone is openly documenting their push toward 5M impressions, they are hunting for engagement. If you reply intelligently to those posts, you get your name in front of an audience that is already highly active and engaged.
The accounts building to 5M impressions are also the ones most likely to be using engagement pods, reply coordination, and mutual amplification. Plug into those networks early, before they become selective.
Verified Accounts in Your Reply Section
Multiple high-performing posts confirm this pattern: getting verified or high-follower accounts to engage your comment section directly increases impression distribution.
Want 1-on-1 Marketing Guidance?
Work directly with operators who have built and sold multiple businesses.
Learn About Galadon GoldOne post with 1,326 likes made it explicit: connecting with active verified users in your comment section causes impressions to increase fast. Another account documented 500,000 impressions directly tied to activity in their comment section.
Creating content that gives large accounts a reason to reply is the move. Controversial takes, specific data, direct callouts, strong opinions - anything that makes a big account want to weigh in. When they do, their followers see the exchange and your post gets routed into new feeds.
The practical move: when you post something designed for reach, identify the two or three large accounts in your niche most likely to have a strong opinion on it, and make sure the post lands in their orbit. Tag them if relevant. Reply to their recent posts with a link or reference. Give them the on-ramp to engage.
Post Links in Replies
X suppresses posts with external links. This is well-documented and confirmed by practitioners across the platform. If your post contains a link to an outside URL, it will get meaningfully fewer impressions than an identical post without one.
The workaround that is working right now: post the tweet without the link, then reply to your own post with the link in the first reply. Your original post circulates without the penalty. Anyone who clicks through to read more finds the link in the thread. You get the distribution of a clean post plus the utility of the link.
This applies to blog posts, lead magnets, YouTube videos, and anything else you are trying to drive traffic to from X. Stop putting links in the body of the post.
Early Engagement Is the Multiplier
X uses early engagement velocity as a core distribution signal. A post that gets 50 likes in its first 30 minutes will be distributed to a much larger audience than one that gets those same 50 likes spread over 24 hours. The algorithm treats early reaction as a quality signal and uses it to decide how far to push a post.
Post when your specific followers are online and ready to engage. The goal is not to post at the "best time of day" in the abstract. The goal is to post when your specific followers are online and ready to engage. Those are different things.
Check your X Analytics for your own peak activity windows. Then schedule posts to land inside that window. When replies come in early - even a handful - the algorithm reads that as signal and starts pushing the post to non-followers in the For You feed.
One additional tactic that accelerates early engagement: reply to your own post within the first ten minutes. Add a follow-up point, a question, a related data point. It creates additional engagement signal and keeps the thread active. You are using the system correctly.
Posting Frequency and the Account Momentum Effect
The algorithm builds a profile of your account over time. How often you post, what engagement you consistently earn, and whether your content drives replies rather than just likes - all of this shapes how much distribution new posts receive before they have proven themselves.
Accounts that post 5 or more times per day tend to see higher average impressions per post than accounts posting once a day, even when the content quality is similar. The reason is that frequent posting trains the algorithm to route your content more aggressively. Your account becomes a known entity to the recommendation system.
This is the same logic that explains why the reply-guy strategy works at scale. Posting 50 replies per day is a form of high-frequency signaling. You are telling the algorithm that your account is active, engaged, and worth surfacing.
Find Your Next Customers
Search millions of B2B contacts by title, industry, and location. Export to CSV in one click.
Try ScraperCity FreeThe risk: if you post frequently and your content consistently earns low engagement, the algorithm learns that too. Volume without quality destroys account momentum. Post at high frequency so that some posts break through and earn real engagement. That creates a rising baseline for everything you post.
What to Stop Doing Right Now
The data makes a few things very clear.
Stop reposting content from other platforms. The 90% impression penalty applies to your whole account, not just the reposted post. One lazy repost can drag down the distribution of everything else you publish.
Stop writing medium-length posts. Tweets in the 140 to 280 character range averaged fewer views than both shorter and longer posts. If you are going to write a paragraph, either cut it to one sharp sentence or expand it into something substantial.
Stop treating threads as your primary format. Threads averaged only 1,353 views in our dataset. They have their place in depth and detail, but they are not impression drivers. They are tools for converting existing followers, not for reaching new ones.
Stop putting links in the body of your posts. Move links to replies.
Stop treating follower count as the primary growth metric. The engagement rate data shows that large follower counts with weak engagement actively hurt impression distribution. Build slower if you have to. Build a responsive audience.
Building a Repeatable Impression System
The accounts compounding impressions month over month are not doing one thing. They are stacking several behaviors that reinforce each other.
They post original talking videos at least once or twice per week. They post 3 to 5 short punchy text posts per day. They reply heavily to accounts in their niche to seed name recognition. They respond to their own early replies within the first 30 minutes to boost engagement velocity. They post links in replies, not in the body of posts. Tracking which post formats earn real engagement for your specific audience and doubling down on those is what separates accounts that plateau from accounts that keep climbing.
The constraints theory principle applies here: your impression growth is only as fast as your weakest variable. If your hooks are weak, better posting frequency will not help. If your posting frequency is low, better content will hit a ceiling. If you are still putting links in post bodies, everything else you do is partially suppressed.
Identify the single thing that is most throttling your impressions right now and fix that first. For most accounts, it is either format (not using video), repost behavior (triggering penalties), or engagement velocity (posting at the wrong time for their audience).
Execution is the difference between accounts that understand these behaviors and accounts that grow from them.
If you are building your X presence as part of a broader growth strategy and want tools that can handle the content side - scheduling, AI-assisted writing, viral format research - Try SocialBoner free for 7 days and see how it fits your workflow.