Am I Shadowbanned on YouTube? The Real Diagnostic Checklist (and the 4 Things It Usually Is Instead)

Short answer: probably not. YouTube has never used the word "shadowban" in official communication. What the platform has publicly described — for years — is reducing recommendations for borderline content and applying limited-ad tiers to videos that don't quite cross policy lines. Both produce shadowban-like symptoms (views drop, recommendations dry up, search visibility softens) without being a hidden account-level penalty. The much more common explanation when your views fall off a cliff is one of four things: a recommendation-engine signal change, a "Made for kids" misclassification, a niche/topic shift you can feel in your comment section before you see it in views, or YouTube's 2026 ranking-signal rebalance toward viewer satisfaction over raw watch time. This article walks you through a real diagnostic checklist using YouTube Studio's Traffic Sources tab — and then explains what your comment section is trying to tell you that the analytics dashboard isn't.

What YouTube has actually said

YouTube's canonical statement on visibility reduction is a December 2019 blog post titled "The Four Rs of Responsibility." It's still the operative framework in 2026; there hasn't been a formal replacement.

The relevant verbatim quotes:

"We've begun reducing recommendations of borderline content or videos that could misinform users in harmful ways."

"The result is a 70% average drop in watch time of this content coming from non-subscribed recommendations in the U.S."

"Since January 2019, YouTube has launched over 30 different changes to reduce recommendations of borderline content."

Notice what's in those quotes and what isn't. The mechanism is reducing recommendations. The trigger is borderline content — content that "comes close to — but doesn't quite cross the line of — violating our Community Guidelines." The scope is non-subscribed recommendations (so your loyal subscribers will still see your videos; the people YouTube would otherwise have introduced you to will not). And the magnitude is concrete: roughly a 70% reduction in that surface.

Nothing about that is "shadowban" the way creators use the word. There's no hidden penalty, no flag on your account, no invisible blacklist. There's a published policy, a measurable mechanism, and a stated outcome. The difference matters because if you think you're shadowbanned you'll spend three weeks Googling "shadowban test" and posting on Reddit; if you understand it as a content-policy signal you can actually look at your last six videos and ask which ones might have brushed the borderline.

YouTube's algorithm team has also been more public about how recommendations actually work. Todd Beaupré (Senior Director, Growth & Discovery) summarized it in a January 2025 interview like this: YouTube pulls videos based on what a viewer is likely to want next; it doesn't push videos because a creator made them. The algorithm is built for the viewer, not for you. That's not a shadowban talking. That's the operating principle.

So before we start the checklist — set the panic aside. There's no secret penalty. There are several real mechanisms that look like one, and one analytics tab that tells you which mechanism you're actually dealing with.

The 60-second diagnostic: Traffic Sources tab

Open YouTube Studio → Analytics → Reach → Traffic source types. This is the single most useful screen in YouTube Studio for diagnosing a view drop. The official categories, in YouTube's own words from their Help documentation:

• Browse features = "Home, subscriptions, Watch Later, Trending/Explore, and other browsing features"
• Suggested videos = "suggestions that appear next to or after other videos"
• YouTube Search = "search results"
• External = "websites and apps that have your YouTube video embedded or linked to"
• Direct or Unknown = "direct URL entry, bookmarks, signed out viewers"

Compare the last 28 days to the previous 28. What dropped, and what stayed the same? Three patterns cover almost every real case:

Pattern A — Browse + Suggested down, Search flat

This is the most common case among creators who think they're shadowbanned. Your audience that already searches for you still finds you. The recommendation surfaces — the algorithm pulling you toward new viewers — softened.

Real causes:

• A recent video brushed the borderline content policy (controversial topic, copyright-borderline music, claim-territory titles)
• A "Made for kids" misclassification (more on this below)
• Audience signal drift — YouTube reads your average view duration, satisfaction-survey responses, and post-watch session contribution to decide whether your channel produces watch experiences viewers want more of
• You changed niche or topic, and the algorithm is recalibrating who to surface you to

This is the case where the comment section starts giving you the answer before the analytics do. We'll come back to that.

Pattern B — Everything down, all surfaces at once

If Browse, Suggested, Search, and External all dropped roughly equally, the issue is almost never an account-level penalty. It's video-level performance.

Real causes:

• Your last 2–3 videos underperformed on retention (audience dropped off in the first 30 seconds), which signals "low watch experience" to the algorithm, which then reduces distribution of subsequent uploads
• Thumbnail or title fatigue — viewers learned to skip your previews
• Niche saturation (your topic suddenly has 40 new competitors uploading the same angle)
• Posting cadence change (you went from 2/week to 1/month and the algorithm deprioritized you for inactivity)

Pattern C — Search down only, Browse + Suggested flat

This is an SEO problem, not an algorithm problem. Your existing distribution is fine; you stopped showing up in searches you used to rank for.

Real causes:

• A title/description change pushed you out of an existing search ranking
• A larger channel covered the same query with a stronger video
• Search intent shifted (a 2024-relevant query is now answered by 2026 content)

The Traffic Sources tab won't tell you why any of these happened. But it will tell you which of them you're dealing with — and that alone narrows the search from "the algorithm hates me" to a specific, fixable thing.

The 4 things it's usually actually about

Once you know which Traffic Source pattern fits, here are the four real causes that account for the overwhelming majority of "shadowban" panic.

1. Borderline-content recommendation reduction (the closest thing to a real "shadowban")

YouTube's Four Rs framework explicitly reduces recommendations for content that "brushes up against" Community Guidelines without violating them. The result, in YouTube's own words: a 70% average drop in non-subscribed recommendation watch time.

The signal pattern is Pattern A above: Browse and Suggested drop hard, Search stays flat, subscribers still watch. If your recent uploads touched controversial framing — politics, health claims, conspiracy-adjacent topics, anything involving minors — this is the likely cause. The fix isn't a shadowban appeal; it's an editorial choice on the next upload.

YouTube has also iterated this system: as of March 2025 they added a human review layer to limited-ad decisions, and in July 2025 they relaxed profanity rules (full monetization is allowed even with strong profanity in the first seven seconds). The policy isn't static.

2. "Made for kids" misclassification

YouTube auto-tags videos as "Made for kids" using content cues. Misclassification strips a video from adult recommendation surfaces. The platform is explicit that creators are legally responsible for setting this correctly — they tell you not to rely on auto-detection.

This is not theoretical. In September 2025 the FTC settled a $10M COPPA case with Disney over misclassification on YouTube. If it can happen to Disney, it happens to creators with characters, toys, animation, family-friendly framing, or any aesthetic that could trip the auto-classifier.

Check the "Made for kids" toggle on your last 10 uploads. If anything looks miscategorized, that alone can explain a recommendation drop.

3. Niche or audience drift the comment section saw first

This is the cause that almost no other "shadowban" article covers. Your view drop isn't an algorithm penalty — it's the algorithm reading a real signal you missed: your audience has been telling you they're losing interest, in the comments, for weeks.

What that actually looks like:

• Comment volume per view declining while view count looks flat. Subscribers are still watching but not engaging. Algorithm reads this as "lower satisfaction" — distribution softens, view count follows.
• Sentiment shift in the comments — from enthusiastic to neutral, or from neutral to critical, on topics you used to dominate. Often the same viewers, with measurably different tone over a 60-day window.
• Topic confusion — comments increasingly contain "wait, did you used to make videos about X?" or "this isn't what I subscribed for." That's the algorithm's least favorite signal: a channel whose audience can't categorize it.
• Recurring questions your audience keeps asking that you haven't answered. If 40 comments across your last 10 videos say variations of the same question and you haven't made that video, the audience is telling you what to do next — and the algorithm is reading your declining ability to deliver what they want.

This is what YouTube comment intelligence is for. The view-count panic is downstream. The signal is upstream, in the comments.

There's a wedge here that almost nobody uses: you can also check your competitors' comment sections for the same drift. If their audience is showing the same engagement decline at the same time, you're dealing with a niche-wide algorithm rebalance and not a personal problem. If their audience is fine and only yours is shifting, the story is about you and your last 20 videos. Same data, opposite conclusions — and the only way to tell them apart is to read someone else's comments.

"If your competitor has 200 comments asking variations of 'but how does this work for small businesses?' — and they're not making that video — you are."

OneTube's free Spy Mode audit at onetube.io/audit does exactly this: paste a competitor channel URL, get a Pulse Report by email. Sentiment trends, intent classification, recurring questions, theme clustering. $0, no card, no account needed. The first report often answers the question "is this me, or is this the whole niche?" — which is the actual question hiding underneath "am I shadowbanned."

4. The 2026 ranking-signal rebalance (the "satisfaction" shift)

The biggest under-reported algorithm change of 2026: the ranking signals YouTube weights most heavily have moved away from raw watch time and toward viewer satisfaction surveys and post-watch session contribution. Translation: the algorithm now cares less about whether someone watched your video for 8 minutes and more about whether they were glad they did, and whether they kept watching YouTube after.

This isn't a single dated announcement; it's a pattern multiple 2026 algorithm summaries have triangulated from official creator communications and Beaupré's 2025 interview. A practical consequence: channels that used to optimize for "longest possible video for max watch time" are seeing recommendation drops while shorter, denser videos hold steady. If your channel's strategy was watch-time-maximalist, that's the cause — not shadowban.

How to tell: pull your average view duration as a percentage of video length over the last 6 months. If percentage retention dropped while absolute watch time held flat, that's the shift biting you. Make shorter videos. Or denser ones. Or both.

The actual "shadowban test" creators should run

Forget the third-party shadowban-checker tools (YTLarge, TubePilot, Bulkoid, Shadowban Detector — they all promise a verdict in a single click and deliver nothing actionable). Run this instead:

Steps 1–5 take about 20 minutes and answer the algorithm-side of the question. Step 6 is where the actual cause usually shows up — and is the part most creators skip because reading 200 comments across two channels by hand is awful.

The third-party "checker" tools — what they actually do

Most of the top SERP results for "am i shadowbanned on youtube" are tools that ask for a channel URL and return a green or red verdict. Here's the honest read on what they're checking:

• YTLarge / TubePilot / Bulkoid / Shadowban Detector — they query whether your videos appear in YouTube search results for their own titles when run against a generic IP. That's it. They don't have access to your Studio analytics, they can't see your traffic sources, they can't read your comments. A "green" verdict means "your videos show up in search," which is also what Pattern A above predicts when you're not shadowbanned but your recommendation surfaces dropped. A "red" verdict often just means the tool's query timed out.
• The "logout/incognito comment test" — open a private window, watch one of your videos, post a test comment, then sign in and check whether the comment shows up in YouTube Studio's "Held for review" tab. If it's held there, that's normal moderation, not shadowban. If it never shows up anywhere, that's worth checking — but it tells you about that one video's comment moderation settings, not about your channel.

Neither of these is useless. Both are vastly less useful than the 6-step checklist above.

Did YouTube change the algorithm in 2026? Yes, but probably not the way you think

Three documented 2026 shifts that show up in creator analytics:

1. Satisfaction surveys + session contribution outweigh watch time as the dominant ranking signal. (Covered above.)
2. Shorts hit 200B daily views (per Neal Mohan's early-2026 CEO letter). Shorts traffic is now a meaningful chunk of total platform attention, which pulls some recommendation surface away from long-form for the same channel.
3. Limited-ads policy iteration — March 2025 added human review to the limiting decision; July 2025 relaxed profanity timing rules. Both reduce the false-positive limiting rate but don't change the underlying mechanism.

What 2026 didn't do: introduce a new account-level penalty, change the Four Rs framework, or shift the recommendation engine away from being viewer-led. If you read a creator forum post claiming "YouTube quietly nuked my channel in the 2026 update," the more likely story is one of the four causes above plus the satisfaction-survey shift.

What to do if it's a real borderline-content issue

If your Traffic Sources show Pattern A and your last 3-5 uploads touched content that could be read as borderline:

• Don't appeal "shadowban." There's nothing to appeal — there's no flag on your account.
• Look at the specific videos. Which one or two are pulling Browse/Suggested down? In Studio, sort by Impressions over the last 28 days; the videos whose impressions dropped most are the ones the algorithm de-surfaced.
• Compare those videos to your earlier hits. What changed? Topic? Framing? Thumbnail aggressiveness? Title claims?
• Adjust next upload accordingly. Recommendation surfaces typically recover within 2–4 uploads if the new content stays on-policy.
• Check limited-ad status. If you see yellow $ icons on the dropped videos, request manual review — YouTube's March 2025 update added a human review layer specifically for these decisions.

The recommendation engine forgives. It doesn't forget instantly, but it forgives.

FAQ

Am I shadowbanned on YouTube?

Almost certainly not. YouTube has never used the word "shadowban" in official communication. What does exist is recommendation reduction for borderline content (publicly documented, 70% average watch-time drop on non-subscribed recommendations) and limited-ad tiers for content that brushes against advertiser-friendly policy. Both produce shadowban-like symptoms without being a hidden account penalty. Run the 6-step checklist above to figure out which mechanism — if any — is actually affecting you.

How can I tell if YouTube shadowbanned me?

Open YouTube Studio → Analytics → Reach → Traffic source types. Compare the last 28 days to the previous 28. If Browse + Suggested both dropped while Search stayed flat, you're looking at a recommendation-surface signal change (borderline content, misclassification, or niche drift). If everything dropped equally, it's video-level performance. If only Search dropped, it's an SEO issue. None of those three patterns require a "shadowban" diagnosis.

What's the best YouTube shadowban test in 2026?

The third-party "shadowban checker" tools (YTLarge, TubePilot, Bulkoid, Shadowban Detector) check whether your videos appear in YouTube search. That tells you almost nothing, because shadowban-like symptoms often leave search visibility intact while killing recommendation surfaces. The real diagnostic is the Traffic Sources tab in YouTube Studio plus a comment-section read on your own channel and 1-2 niche peers. The free OneTube Spy Mode audit handles the comment-section side at $0.

Why does the YouTube algorithm hate me?

It doesn't. The algorithm doesn't form opinions about creators. It optimizes for viewers — surfacing videos a given viewer is likely to want next, based on watch history, satisfaction surveys, and session contribution. If your distribution dropped, the algorithm read a signal — about borderline content, about audience drift, about retention — and reduced your recommendation share. The fix is figuring out which signal, then changing what you upload next. There's no personal vendetta inside the algorithm.

Did YouTube change the algorithm in 2026?

Yes — the most significant shift is that viewer satisfaction surveys and post-watch session contribution now outweigh raw watch time as primary ranking signals. Channels that optimized for "longest possible video for max watch time" are seeing recommendation drops; shorter, denser videos are holding steady. The shift is gradual, not a single announcement, and it's been triangulated from official creator communications and Todd Beaupré's January 2025 interview rather than a dated blog post.

How long does a "shadowban" recovery take?

If the cause was recommendation reduction for a single borderline upload, recovery typically takes 2–4 subsequent on-policy uploads. If the cause was "Made for kids" misclassification, correcting the toggle fixes it within days. If the cause was niche drift or the 2026 satisfaction-shift, recovery is slower and requires an actual change in what you upload — there's no flag to flip.

Does OneTube detect shadowbans?

No. There's nothing to detect at the account level — shadowban as a discrete account-level event doesn't exist on YouTube. What OneTube does is read your channel's and your competitors' comment sections for the audience-side signals (sentiment drift, recurring questions, topic confusion) that usually show up weeks before the view-count drop hits Studio. Spy Mode is read-only intelligence: we surface the signal, you act on it inside YouTube Studio.

What to do next

• Just want to run the diagnostic now? Open YouTube Studio → Analytics → Reach → Traffic source types. Match your pattern to A, B, or C above.
• Pretty sure it's a "Made for kids" misclassification? Check the audience setting on your last 10 uploads in the Content tab.
• Suspect a borderline-content limit? Look for yellow $ icons in the Revenue column. If you see them on videos you think shouldn't have been limited, request manual review (YouTube added a human review layer in March 2025).
• Want to know if it's you or the whole niche? Read the comment sections of 1-2 channels in your niche over the last 60 days. The DIY version takes about an hour per channel. The fast version: paste any competitor channel URL at onetube.io/audit. $0, no card, no account needed. The Pulse Report by email tells you what their audience is asking for and whether engagement is dropping there too.

Related reads

• YouTube Comments Scraper 2026: 6 Real Methods — what's in the comment section, and how to read it at scale
• YouTube Comment Tracker 2026: 8 Tools, and the One Gap None of Them Fill — for monitoring comment activity continuously, on your channel and others
• YouTube Competitor Analysis: the cheapest content brief you'll ever get — when the "is this me or the niche" question gets serious
• What Do YouTube Viewers Want? — the audience-signal side of the same diagnostic story