Faceless Niche Finder
Answer four honest questions and get niches that survive your actual constraints — not a list of whatever is popular.
Tell it what you know and press Find My Niches.
Three to five niches, with the first videos and the checks to run.
Working through what fits…
This usually takes ten to twenty seconds.
Picked one? Draft the first script.
Write a script →Why there are no RPM or competition numbers here
Every other niche tool hands you a table: RPM, search volume, competition score, difficulty rating. None of those can be produced by a language model. It has no ad auction data, no YouTube search index and no view counts, so every figure in that table is invented — and it reads exactly like research, which is what makes it dangerous.
That matters more here than anywhere else on this site. A person choosing a niche is committing months of their life to the answer. So the schema behind this tool has no numeric field at all, and the model is told it may not claim a niche is trending, saturated, growing or lucrative. What it returns instead is the part it can genuinely do: reason from what you told it, and hand you the checks to run yourself.
The four questions, and why they are the four
What you know or care about does most of the work. Vague in, vague out — "I like history" produces a generic answer, while "I worked in logistics for six years and I am good at explaining complicated systems simply" produces something that could only have been written for you.
Time and budget are the two constraints that quietly kill channels. A research-heavy niche at four hours a week is not a slower version of the same plan; it is a different plan. And narration changes the answer more than most niche advice admits: a niche that works with a calm synthetic voice reading explanatory copy is often a poor fit for a channel carried by personality, and the reverse holds too.
Read the hard part first
Every niche that comes back is workable. The real question is which one you will still be making videos in six months from now, and that is usually decided by the thing that makes it tiring rather than the thing that makes it attractive. So each niche states, plainly, why people quit it. That section is not there to discourage you — it is there so the objection arrives in week one rather than month four.
Then go and check
The checks are the deliverable. Each one names a phrase to search, where to search it, what to filter by, and what result should encourage or discourage you. If every result is a large channel answering the question completely, that niche is already served. If small channels are getting views on it this year, that is a signal worth having — and it is yours, measured, rather than a number somebody made up for you.
Where it sits in the workflow
This is step one. Once you have picked something, the Hook Generator and the Script Generator turn one of the video ideas into a draft, and the cost calculator tells you what making it every week would actually cost before you commit to the schedule.
Common questions
Why does it not tell me which niche pays best?
Because nobody can tell you that honestly. Ad rates vary by country, season, audience and advertiser demand, and no language model has access to any of it. A tool that prints an RPM figure for a niche is guessing, and the guess reads as research. What this returns instead is the structural money question — whether a niche fits sponsors, affiliates, your own product, or is purely ad-dependent — which is answerable without inventing anything.
Where are the competition and search volume scores?
Deliberately absent. There is no search index, no view count and no ad auction data behind this tool, so any number in those columns would be fabricated. Instead every niche comes with checks you run yourself — specific phrases to search, what to filter by, and what result should encourage or discourage you. Ten minutes of that gives you evidence rather than a number.
How specific are the niches?
Specific enough to be a channel. Not "finance" or "history", but something you could describe in one sentence and start this week. If the suggestions come back too broad, put more of yourself into the first field — what you already know, what you have read about for years, what you would still find interesting on video forty.
Does it matter whether I use my own voice?
More than almost anything else you tell it. A niche that works with a synthetic voice reading calm explanatory narration is often a poor fit for a channel where personality carries the video, and the reverse is just as true. It is a question most niche advice skips, which is why it is one of the four here.
Does it use one of my daily AI generations?
Yes. The free allowance is shared across every AI tool on the site rather than given separately to each one, because each generation costs real money whichever tool spends it. Everything that runs in your browser — the calculators, the thumbnail tools, the converters — stays unlimited.