A decade on Twitter / X: from 140 characters to shadowbans

This year marks my 10th anniversary on Twitter or X, as we are now supposed to call it. Over the past decade, I have watched the platform evolve dramatically, and honestly? Not much of that change has been for the better.

During these 10 years, my account has been suspended twice. One of those times left me locked out for nearly six months. Every time, I filed an appeal, and every time, my account was eventually restored because the suspensions were completely unjustified. I suspect jealous mass reports played a big role. Unfortunately, the damage was done: those suspensions triggered a shadowban that I am still suffering from today, with no clear path to appeal it.

What does being shadowbanned actually mean?

A shadowban is a form of stealth moderation where the platform suppresses your account or content without ever telling you. From your side, everything feels normal - you can tweet, like, and comment just fine. But behind the scenes, your posts are quietly hidden or pushed to the bottom of the pile for everyone else.

On X, shadowbans usually show up in four distinct ways:

• Search Suggestion Ban: Your handle stops auto-completing in the search bar when people type your name.
• Search Ban: Your posts and replies disappear entirely from search results, even if someone searches your exact username or hashtag.
• Reply Deboosting ("Ghost Ban"): Your replies are buried at the very bottom under a "Show more replies" button or hidden behind a content warning.
• "For You" / Recommendation Ban: Your content stops appearing in non-followers' feeds, completely cutting off your organic reach.

A decade of upheaval: the major milestones

Looking back over 10 years, there was really only one major update that felt like a clear win: back in November 2017, when Twitter doubled its iconic character limit from 140 to 280. It finally gave us room to breathe.

After that, the timeline became a roller coaster:

• July 2018: "The Great Purge" wiped out tens of millions of locked and fake accounts, causing follower counts to plummet overnight platform-wide.
• October 2022: Elon Musk acquired Twitter, took full control, fired top executives, and laid off roughly half the staff in sweeping cuts.
• March 2023: Paid blue checkmarks launched via Twitter Premium in Belgium (where I live), and legacy verification badges were slowly phased out.
• July 2023: Twitter officially rebranded to X, replacing the iconic blue bird logo, while temporary view limits were slapped on unverified accounts.
• July 2025: Driven by stricter privacy regulations like the UK Online Safety Act and EU Digital Services Act, X rolled out aggressive age-assurance prompts. Millions of users in regulated areas suddenly hit wall after wall with the notice: "Due to local laws, we are temporarily restricting access to this content until X estimates your age."

That 2025 age verification rollout hit my impression numbers harder than anything else. A huge portion of my community and followers prefer to stay anonymous online and share as little personal data with X as possible. Asking them to hand over identifying details just to see a post turned many of them away.

How X expects you to verify your age

In theory, age verification is straightforward. Usually, simply confirming your birth date in your account settings is enough. However, if X flags your account for extra verification - especially to view sensitive content - you get handed off to a third-party service like Persona. There, you have to upload a photo of an official ID (passport, driver's license, or national ID) and complete a quick facial selfie scan. While approval only takes a few minutes, there is a catch: if you update your birth date to a year showing you were under 13 when you created the account, X will instantly lock you out.

The 2026 algorithm overhaul: a whole new game

If the age gates weren't enough, 2026 brought a total overhaul of X's recommendation engine. The platform scrapped its old legacy code and replaced it with xAI's language models, including Grok. Instead of relying on basic metadata, the algorithm now evaluates content using deep semantic processing. In short: overall post structure, context, and actual writing quality matter far more than quick growth hacks.

Here is how the dynamic has shifted recently:

• Passive likes are dead: Simple likes carry very little weight now. The algorithm heavily favors active, multi-turn conversations. A post that triggers real back-and-forth discussions in the reply section is rewarded with massive reach.
• Engagement velocity is king: The first 15 to 30 minutes after you post serve as a test run. Quick interactions signal high value and push your post into the "For You" feed; without early traction, your post stalls immediately.
• Hashtags are officially obsolete: Because the AI categorizes content semantically, adding hashtags adds zero value. In fact, using three or more can trigger spam filters and hurt your reach. Clear writing is all you need.

Where does that leave us?

For the past few months, my feed has felt broken. I barely see the accounts I actually choose to follow in my "For You" timeline. Even on the dedicated "Following" tab, smaller creators have vanished, replaced mostly by high-volume engagement-bait accounts.

On my own profile, impressions and engagement have dropped to all-time lows. It feels like my posts are being broadcast into an empty room, never even reaching the people who signed up to see them.

It makes me wonder why I bother staying. Do I post for myself? Not really. I post for my community - but if the platform prevents them from seeing my work, what is the point?

Bluesky is looking like a refreshing alternative these days. It has its own set of flaws, sure, but that is a topic for another blog post.