Is my niche too broad for the algorithm?

Breadth is not saturation, and the two get confused constantly

They produce the same complaint and need opposite fixes.

Saturation is competition density — too many people making the thing you make, so the thing you make is hard to stand out inside. The fix is a sharper angle or an underserved corner. We cover that question per niche, for example on is the gaming niche too saturated.

Breadth is identity clarity — nobody, including the recommendation system and including your own followers, can predict what your next post will be. The fix is coherence, and it is not the same as narrowing.

If your posts are all in one topic and none of them travel, that is a saturation-or-execution question. If your posts are scattered and the good ones do not lift the ones after them, that is this page.

The mechanic, without mythology

Neither platform publishes a niche score, and no setting exists that you have accidentally tripped. What both describe is that a post is offered first to a small group most likely to want it, and how that group responds decides whether it is offered more widely.

Your followers are the cheapest and earliest such group. That is the whole mechanism by which breadth costs you something:

• You post a cooking video to an audience that came for gaming.
• Your own followers scroll past — not because it is bad, but because it is not what they are here for.
• The earliest signal available is therefore weak, and the post does not widen.
• You conclude the cooking video failed. What actually failed was testing it on the wrong people.

Notice what this implies. Breadth is not punished in the abstract. It costs you precisely at the point where your existing audience is the wrong first audience for a given post — which also explains why the same video can do fine on an account with no established expectation at all.

The test, on your own last twenty posts

Ten minutes, no tools, and it answers the question better than any rule about niches.

Open your last twenty posts. For each one, ask: a person who liked this post — would they also like at least three others in this list?

• If nearly every post has three or more companions, you have one channel. Breadth is not your problem, whatever the topic labels say.
• If most posts have zero or one, you have several small channels sharing a login. Each post is starting the audience-finding problem again from scratch, and none of them compound.

Then the confirming read in your own analytics. For your best and worst recent posts, compare the share of views from people who already follow you versus people who do not. Wildly different splits per post is the fingerprint of an account whose audience does not carry from one upload to the next.

The thing that has to be consistent is not the topic

This is the part the standard advice gets wrong, and it matters because "niche down" is often the wrong prescription.

Plenty of large accounts cover enormous topical range. What they never vary is the reason someone comes back, and that reason can be any one of three things:

1. A topic. The most obvious and the easiest to sustain. Everything is about one subject.
2. A format. The subject changes; the shape does not. Same structure, same length, same kind of payoff every time. This is how variety shows work.
3. A voice or a person. People are here for you. This is the hardest to build and the most durable once built, and it is the only one that genuinely licenses topic-hopping.

You need one of the three to be legible from three consecutive posts. If a stranger could scroll three of your posts and not name any of them, that is the actual finding — and it is more useful than a verdict on whether your niche is "too broad".

"My channel is all over the place — how do I figure out my niche?"

Do not pick a niche in the abstract. Read it off your own results, which is both faster and more likely to be something you can sustain.

1. Sort your posts by performance and look at the top five. Ignore the ones you are proud of; look at the ones that travelled.
2. Write one sentence describing what those five have in common that is not "they are good". A subject, an audience, a format, a promise.
3. If nothing is in common, look at the second thing they share — the ones that recruited followers rather than the ones that got views. Studio and Insights both expose this, and it is the more useful column for this question.
4. Make five posts against that sentence, deliberately, before deciding whether it is right.

Two accompanying rules. Give a direction enough posts to be judged as a median rather than as a single result — the variance between posts on the same account is large enough that one upload proves nothing, which is unpacked in why one video blows up and the next one dies. And if you change direction on an established account, expect a slow patch while the audience re-sorts, which is the cost of the change rather than evidence it was wrong. Whether that cost is worth paying at all is the subject of start a new channel or fix the one you have.

"I've tried multiple niches and still get almost zero impressions"

This one needs separating, because it is usually two problems stacked and only one of them is about breadth.

Zero impressions means you are not being shown, which is upstream of everything on this page — no amount of topical coherence matters if distribution never starts. Diagnose that first with zero impressions versus zero views, which separates never-shown from shown-and-passed-over.

The breadth part is the pattern of repeated switching itself. Each switch restarts the audience-finding problem, so a creator who changes direction every few posts never accumulates the thing that makes any direction work. It is the worst of both worlds: none of the coherence of a narrow account, and none of the personality-driven latitude of a broad one. If you have tried several niches, the most valuable next move is usually to pick the least-bad one and hold it long enough to produce a median.

What our own data can and cannot tell you here

We publish a 360-channel dataset drawn from public YouTube channel statistics, and it is worth saying plainly what it does not contain: no topic or niche field. It cannot measure whether varied channels underperform focused ones, and we are not going to imply otherwise by dressing up an adjacent number.

What it does show is relevant in one narrow way. Views per video as a share of subscriber count runs from 0.1% at the 5th percentile to 1,147.5% at the 95th across those 360 channels, with a median of 50.2%. The caveats are published with the figure — these are lifetime averages, the sample is channels whose owners sought an audit rather than a random draw, 41.7% are in India, and there is no channel-age field. The relevant point for this page is narrow: outcomes at a given size are so widely spread that no single-axis rule, breadth included, is going to explain your account. The full methodology is on small YouTube channel statistics.

Where a tool helps, honestly

Disclosed interest: we build a diagnostic, so treat this as an interested party being specific rather than as neutral advice.

The twenty-post test above is the real work and you can do it for free in an afternoon. What is genuinely hard alone is being honest about your own catalogue — you remember what each post was meant to be, which is exactly the memory that stops you seeing what it actually was. Channel X-Ray reads your recent uploads and names one bottleneck, including when the bottleneck is that your posts are not building on each other. The free YouTube channel audit and free Instagram Reel analyzer give a lighter read with no signup.

For the outward question — what is being watched in a topic you are considering — Viral Radar searches across other creators. Be aware of what that is and is not: it is topic search across other people's videos rather than a read of your own history, and it costs credits after the signup grant.