Am I Shadowbanned? The 2026 Diagnostic That Actually Answers It

Neither TikTok nor Instagram has ever shipped a feature called a shadowban — and both have shipped, publicly and by name, several features that do exactly what creators mean by the word. That contradiction is why the question is so hard to answer honestly, and why “you’re not shadowbanned, it’s just the algorithm” is such an unsatisfying reply.

Here’s the direct answer. “Shadowban” is a symptom, not a mechanism. Three completely different things produce it: a single video ruled ineligible for recommendation, an account-level restriction on recommendations, and removal from search. Each one has a different test, a different timeline, and a different fix — and the fourth possibility, by far the most common, is that nothing was restricted at all and your content simply got reclassified.

Twenty minutes of structured checking will tell you which of the four you’re in. Guessing will not, and the guesses are expensive: the standard panic response is to delete posts, strip captions, and change everything at once, which destroys the evidence you’d need to diagnose it and resets the very signals you’re trying to recover.

What the platforms actually admit to

Start with the vocabulary, because the official language is the diagnostic.

TikTok denies “shadowbanning” as a policy while maintaining a published distinction between content that is allowed on the platform and content that is eligible for the For You feed. A video can be perfectly within the community guidelines and still be held out of recommendations — because recommendation eligibility is a separate, stricter bar. TikTok exposes this per-video in the app.

Instagram is more explicit. It has publicly stated that it limits how widely it recommends content that doesn’t meet its Recommendation Guidelines, and it built Account Status specifically so creators can see whether their account or a given post is affected. Head of Instagram Adam Mosseri has repeatedly framed the goal as making this visible rather than denying it exists.

So the honest framing is this: no platform quietly bans you without a signal, but both platforms make eligibility decisions that are much finer-grained than “banned or not.” The creator-facing surfaces are real. Almost nobody checks them before posting the “am I shadowbanned?” question to Reddit.

The three levels of suppression

Level 1 — video-level ineligibility

One post is held out of recommendations. Everything else on the account performs normally.

How it looks: the affected video lands at a fraction of your usual reach — often flat for its entire life rather than rising and stalling — while the post before it and after it behave exactly as expected.

The test: open the video’s own analytics and look at traffic sources, not the view count. On a healthy video, the For You feed or Explore is the dominant source. On a video-level restriction, that source is near zero while “personal profile,” “following,” and “search” still show normal-looking numbers. The video reached the people who came looking for it and was never handed to anyone else.

The usual causes are mundane: music you don’t have rights to in some regions, a claim on a clip you re-used, borderline content that tripped an automated review, or a caption that reads as a policy edge case. Both apps will tell you the status of the specific post if you ask — TikTok in the video’s own menu, Instagram under Account Status.

Level 2 — account-level restriction

Every post since a particular date underperforms, together.

How it looks: a cliff, not a slope. The distinguishing feature is that the drop lands on all content at once, including formats that reliably worked for you before, and it starts on a specific day rather than drifting down over three weeks.

The test: check Account Status first — this is the one level where the platform will usually just tell you. Then check whether your follower-sourced views fell too. This is the single most useful distinction in the whole diagnostic. A recommendation restriction cuts non-follower distribution and leaves your existing audience largely intact. If your followers are still watching at the usual rate but nobody new is arriving, that’s consistent with a restriction. If reach fell across both, you’re more likely looking at a content or classification change.

Level 3 — search and discoverability

Your videos still get recommended, but you can’t be found on purpose.

How it looks: normal-ish reach, but your handle doesn’t autocomplete, your posts don’t surface for terms that clearly describe them, and inbound traffic from search flatlines while the feed keeps delivering.

The test — the unique-tag check. Post a video with a nonsense string in the caption that exists nowhere else on the platform (something like qzvrmk plus four digits). Wait an hour. Then, from a logged-out browser or a second account that doesn’t follow you, search that exact string. If the video doesn’t appear, your content isn’t being indexed for search. Repeat the same check with your handle and with one obvious topic phrase from the video.

The logged-out part is not optional. Searching from your own account is the single most common way creators convince themselves they’re fine — the app cheerfully shows you your own content regardless. That’s also why “my friend can see my post” proves nothing if your friend follows you.

Search suppression matters more than it used to. In-app search has become a genuine second discovery engine, and being invisible there is a slow leak rather than a dramatic drop — which is exactly why it goes undiagnosed for months. We covered how that surface works in TikTok SEO in 2026.

Level 4: it isn’t a ban, it’s a classification change

This is the outcome in most of the cases we see, and it’s the one nobody wants to hear, because “you weren’t punished, you were re-sorted” is harder to act on than “you were punished.”

Both platforms decide who to show a video to before they decide how many people to show it to. That decision runs on what the content appears to be about, and it is revised continuously. When it changes, your reach changes — with no violation, no notice, and nothing to appeal, because nothing went wrong.

Four common triggers:

• A niche drift. Post three videos outside your usual subject and the system’s model of your audience gets muddier, so the first test audience for post four is a worse fit. Reach falls, and it feels like punishment. The mechanics are in how Instagram decides what niche you’re in.
• A new account still calibrating. Accounts under a few weeks old are being sized up, and a flat ceiling during that window is the system working as designed, not a penalty. See why your views are stuck at 200.
• A post-viral baseline. After a hit, your next video is measured against an inflated expectation and a temporarily mismatched audience. It underperforms relative to the spike while being completely normal. That’s the viral hangover.
• A reading error. You’re comparing raw view counts across weeks with different distribution, which mostly measures the weeks. Ratios survive that comparison; raw counts don’t. We listed the ones that actually predict next month in which metrics matter.

And one myth worth retiring: the belief that certain hashtags get you flagged. Tags are topic hints. A wildly off-topic or spammy tag stack can make your content harder to classify, which costs you reach — but it isn’t a ban, and no tag list is a landmine map. The full version is in what hashtags actually do now.

The 20-minute diagnostic

Run these in order. Stop at the first one that answers the question.

1. Check Account Status (Instagram) or your video’s status and account eligibility (TikTok). If either flags something, you have your answer and an appeal path. Most people skip this step entirely.
2. Pull traffic sources for the last 10 posts. Note the share coming from the recommendation feed. A collapse in that one source with search and profile intact is the fingerprint of a recommendation restriction.
3. Split follower vs non-follower reach. Followers holding while new reach vanishes points to restriction. Both falling points to content or classification.
4. Run the unique-tag search test, logged out. This is the only reliable read on search indexing.
5. Plot the drop. Cliff on a single date across all formats = restriction. Gradual slide = classification, format fatigue, or a real change in your content quality.
6. Look at what changed in the seven days before the drop. New audio source, a subject shift, a link in bio, a caption style, a burst of posting. Something almost always did.

If it is real — and if it isn’t

If a level flags: appeal in the app and then wait. Restrictions of this kind are typically time-boxed rather than permanent, and the recovery behaviour that works is boring — keep posting on-niche at your normal cadence, remove the specific offending element, and don’t touch anything else so you can tell what fixed it.

Do not delete the affected posts. Deletion removes the evidence and, on an account already being re-evaluated, a burst of deletions is not a helpful signal. Do not go silent for two weeks; a gap makes the classification problem worse whether or not there was ever a restriction. Do not change your format, posting time, caption style, and hashtag strategy in the same week — that guarantees you learn nothing.

If nothing flags — which is the likely outcome — then the honest conclusion is that your reach problem is a content problem, and that’s better news than a ban. Bans expire on someone else’s schedule. Content you control.

---

The reason the shadowban question is so persistent is that creators only get feedback after publishing, when the only available move is interpretation. A video that gets held back and a video that simply didn’t hold attention look identical from the outside, days too late to do anything about it.

That’s the part Blossom changes. It reads your video before it goes out — scoring the hook, the pacing, and where attention is likely to drop — so the question you’re answering is “is this strong enough to earn distribution?” rather than “why was I punished?” Run a breakdown on your next post, or read the short answers in the FAQ.

The word “shadowban” describes a feeling. Traffic sources describe a fact. Check the fact first.