Why X Posts Are Getting Hidden After the Mid‑May Update — And How Creators Can Avoid It

By CaylaCatz

Why X Posts Are Getting Hidden After the Mid‑May Update — And How Creators Can Avoid It by CaylaCatz If you’ve noticed your posts on X suddenly getting fewer views, disappearing from timelines, or being “hidden” when you tag multiple people, you’re not alone. Since mid‑May, many creators — especially AI artists, poets, and small collaborative communities — have been reporting the same thing. The same thing happened to me. Because it was driving me crazy, I explained to copilot about what was happening to me and my friends who like to share AI art through QTs but suddenly in mid-May we are being intermittently shadowbanned. People don't see our post in the feed. X tells everyone it's "unavailable" but people can see replies to the post. So weird! omg! Copilot helped me to see by confirming what my friends and I have been noticing but also suggested some alternative strategies so we can still share our art and build our communities. This blog breaks down: 🔘 what changed 🔘 why tagging 3–5 people now triggers suppression 🔘 how to collaborate safely 🔘 how to keep your reach strong without risking a shadowban These are community observations , not official documentation. X has not shared the changes they've made publicly. But the patterns are consistent enough that creators can adapt and protect their visibility. 🚨 What Changed in X’s Algorithm (Mid‑May) Around mid‑May, X quietly adjusted its algorithm to crack down on what it calls “coordinated posting behavior.” Unfortunately, this change affects normal, harmless creative collaboration like our QTs. Creators began noticing: 🔘 posts with 3–5 tagged users sometimes get hidden 🔘 QT chains get suppressed 🔘 posts vanish from timelines for hours 🔘 engagement drops sharply after tagging groups 🔘 replies stay visible, but main posts disappear This isn’t targeted at artists — it’s an automated system trying to detect spam, bot rings, and engagement farming. But our small creative communities get caught in the crossfire and for me anyways, I took it personally. I had just started X and had two wonderful weeks of sharing and then bam! People couldn't see me. At first I thought it was because I was new, had few followers and was on the free tier. Turns out, it was none of the above. It was a change in the algorithm that they announced in mid-May but they didn't spell out what the changes are. We have to figure it out the best we can. ⚠️ Why Tagging 3–5 People Now Triggers Suppression The new system flags posts that look like: mass tagging repeated tagging of the same group or same people burst QTs coordinated chains rapid‑fire replies multiple similar posts in a short window Even if you’re just inviting friends to join a theme, the algorithm may interpret it as spam‑like behavior. This is why posts sometimes get hidden only when you tag multiple people — and why it doesn’t happen every time. The detection is inconsistent. 📌 The Good News: You Can Still Collaborate You don’t need to stop doing QTs, themes, or community art threads. You just need to adjust how you do them. Below are shadowban‑safe templates and safe QT strategies that creators across multiple communities have tested successfully. 🛡️ Shadowban‑Safe Collaboration Template This template avoids every known trigger: no mass tagging in the main post no repetitive patterns no burst‑posting no spam signals no coordinated‑behavior flags 1. Main Post (Safe) Use this structure: (Note: use the basic structure but change up the words so no two QTs are the same. If you copy and paste the same phrases for your QT all the time, it can be seen as bot behavior. Remember: Variety is the spice of life. Changing your words up is a safety strategy to separate you from bots. ) Today’s piece: [title or theme]. Open invite — if this inspires you, feel free to join in with your own version or interpretation. I’ll be checking replies and QTs throughout the day. Why it’s safe: no tags no spam signals “open invite” is allowed algorithm sees it as organic 2. Reply 1 (Safe Tagging) Immediately reply to your own post: (again vary the words up) Sharing this with a couple friends who might enjoy the theme today: @A @B Why it’s safe: tagging in replies is low‑risk 1–2 tags is below the threshold avoids main‑post suppression 3. Reply 2 (Optional) 10–20 minutes later: (don't forget to pull out your thesaurus or ask copilot/chatgpt/etc for different ways to say this same thing so you say vary your reply with your QTs. and if you're multilingual, that's another way to introduce variety) And if anyone else wants to join, the thread is open. I love seeing everyone’s interpretations. Why it’s safe: keeps the thread active no tags encourages organic participation 4. Reply 3 (Engagement Booster) After someone joins: (yeah, you know what I'm going to say. Variety is the spice of life) These are fantastic — thank you for adding your version! Why it’s safe: boosts visibility no tagging no spam signals 🔁 Safe QT Strategy (Post–May Algorithm) This is the part that changed the most. 1. QT only ONE person at a time Never QT 3–5 people in a row anymore. Safe pattern: QT one person wait 10–20 minutes QT another wait again This avoids “burst behavior.” 2. Vary your QT captions Rotate various QT captions. For example: “Joining the theme with my piece today.” “Loved this idea — here’s my version.” “Inspired by your post, here’s mine.” “Adding mine to the thread.” Repetitive captions get flagged. 3. Never tag more than 1–2 people in a QT If you want to invite more: tag them in a reply to your QT or DM them or use an open invite 4. Let THEM tag YOU This is the safest and most powerful trick. When others tag you: it doesn’t count against your spam score it boosts your visibility it creates natural engagement clusters Encourage this by saying: If you join the theme, feel free to tag me — I’d love to see your version. A quick note of caution: While it’s true that letting others tag you is safer than tagging multiple people yourself, you still want to avoid creating a repetitive pattern for your friends. If the same person tags you every single day, the algorithm may interpret their behavior as coordinated or spam‑like, even if it’s innocent. To keep everyone safe, aim for broad‑based tagging instead of the same pairings over and over. For example, one friend might tag you today, but then not tag you again for a few days. Another friend can tag you tomorrow. It’s perfectly fine for you to be tagged daily — the risk is only when the same person repeats the same tagging pattern too frequently . Friends can also participate without tagging by using a quote post , or by dropping a link to your post in a reply to their own QT. These variations break up the pattern and keep everyone’s accounts healthy while still supporting the theme and each other. 5. Avoid tagging the same group or the same people every day To keep you and your friends safe, make sure you rotate who you tag or invite. Even when tagging is done safely (in replies, not the main post), repeating the same tagging pattern day after day can still look automated from the algorithm’s perspective. Think of it as spreading the love around: Tag one or two friends today, choose different friends tomorrow, and skip tagging entirely some days. This keeps your account healthy and prevents you and your friends from being pulled into a repeated pattern that could suppress their posts. You can also use the alternative methods mentioned in Step 4 — such as doing a quote post or dropping a link to your post in a reply — to keep a QT theme going without tagging the same people repeatedly. These variations let you support your community while avoiding patterns that might trigger suppression. 6. Don’t QT more than 4–5 people per day I used to invite 4-5 people in one QT. Not any more. Even spaced out, too many QTs can trigger suppression. And that's a royal pain. So now it's max 4 or 5 within a day but spread out so they are not all in the same hour. 🌟 Summary: The Safe Collaboration Formula Main post: no tags open invite unique caption Reply 1: tag 1–2 people Reply 2: open invite again QT others: one at a time spaced out varied captions no more than 1–2 tags Engage: comment on others encourage them to tag you everybody rotate tags so no one tags the same people daily This avoids every shadowban trigger we’ve seen since mid‑May. These strategies have worked for other people so I'm implementing them in how I share my AI art in X posts daily while I continue to be supportive of my AI art friends and community. Let me know if this was helpful for you (I hope so). If you use some of these strategies, let me know if it helps to improve your post-mid-May QT experience. Thanks for reading.

Comments (9)

DavidP

I'm not an X user but still found this interesting. It made me think about our own QT practices and the challenges with notifications we discussed previously. And given that I can see how some of what we consider sound behaviour could be deemed to be 'engagement farming'. Here in BP I guess it is 'farming' to maximise views with hope of likes to achieve 'likes received' goals to earn the 150 credits. So in some respects I can understand the need and intention of this X algorithm/bot. What I do wish is that people would be transparent with users. If introducing such an algorithm, then tell everybody. If you genenuinley believe it is well intended to improve experience or resolve issues like those you described early in the pice then be open about it. As for the good behaviours. I felt a tinge of guilt as I do some of the unsafe practices here when I publish the few posts I make daily. And it is done to try and get the 50 likes daily and is succesful almost all the time. What varies is how many likes I receive which has quite some range, from just over the 50 to a couple of hundred (not sure why). But I still suffer from a flood of mention notifications. And many 'QTs' which I will put in quotes as I don't think they are in the spirit of QT (as I see it). Might as well leave off the label and just tag. And having QTed a topic, probably not necessary to tag a number of your own posts on the same topic? I'm going to stop before I rant.

CaylaCatz

yeah, it's tricky. I think it's about visibility. The main QT list here is long. On X, at most people at QT inviting 5 people. I have found here that if I QT 1 main list, i can usually meet my target. If I don't do the BP main QT, then it takes 2-3 posts of selective QT invites of 5-9 people. I think it has something to do with how many people like you within the first 15 minutes of posting. I say first 15 minutes, becasue I've tested adding QTs later on BP as I add them in comments. The next day you might as well do no QT; even 1-2 hours later you will probably only get 10 or so likes. No QT can be 3-7 for me -- 7 only if I spend a fair bit of time visiting folks. I prefer a curated list but then I have to warn people not to copy it as it was the main QT list. --- X is different because the way the algorithm is they don't want you to notify anyone in your description. But the way notification works is different too. If your post description mentions 4 people for example, then everytime someone replies to that post, all 4 people get the reply in their notification. Which I thought was a bit much at first but I got to like it because it allowed people to have group conversations and you can see the conversations without going to the post. In essence the comment area becomes a mini chatroom for all the people listed in the description. So it's not just more engagement because of likes -- it's more engagement because people chat more to other people in their circle under the post. So by punishing people by making their image invisible so people can't see the original post, these mini chatrooms either don't happen or they happen in the reply section under one of the replies. for example Sealine and I both get tagged by your QT. Sealine saw your picture and replied. I see Sealine's reply but when I click to see the original post -- your post, I can't. Your post is blocked to me although Sealine could see it. So all the chats starts to happen under Sealine's reply.

Unleashed

Ironically, I sit next to a person who has an issue with this daily, but, to my knowledge, I don't. I have in the past, but not currently. And as @⊱❀⊰ Sealine ⊱❀⊰ pointed out to me...I'm doing everything wrong. I do almost everything your blog says NOT to do. So while I agree that your suggestions might help someone having the issue, I also think the algorithm is fickle, and doesn't affect everyone the same. And based on that, I assume my time is coming. I do plan to use the "Safe" and "Unsafe" graphic though, and start making some changes. The blog is very informative, and is accurate for everything I read when I was looking for solutions for Sealine's issues.

DavidP

Thanks for more details. And I think things are excerbated on BP by all the prolific posters who tag everybody on everything. I think either have to stop worrying about the likes reward and just post with a small groups of people you are most connected to or you have to join the band of high volume taggers. And that of course adds to the problem and makes people more inclined to tag more, post more etc etc etc. I saw your tag on the QT haiku which made me visit here to check if you'd replied. No idea if I was notified as thgere are just so many.

chris70

Thank you very much for this comprehensive summary. I will take this into account in the future. And feel free to tag me in your posts every now and then. ☺️

Unleashed

Since I posted the above, I'm getting hit pretty hard. Several of mine have been blocked in just a couple days, as verified by other users.

Sealine

Will definitely try some of these steps and see if it helps. Thanks for putting this all together 😊