Choosing a faceless niche you can still be publishing in six months

Almost every guide to picking a faceless niche ranks niches by ad rate, hands you the list, and stops. Finance at the top, gaming near the bottom, pick from the top.

It is not useless information. It is just the least important of the three variables that decide whether a channel works, and leading with it is how people end up abandoning a finance channel after nine videos.

The three variables are what a video earns, what a video costs to make, and whether you will still be making them in six months. RPM only covers the first.

Why RPM is the wrong place to start

Ad rate varies enormously by topic. Finance, business software and insurance sit at the top; entertainment, compilations and general vlogs sit near the bottom. The gap between them is real and can be several times over.

But RPM only converts views into money. It says nothing about how hard the views are to get, and the two are inversely related for an obvious reason: everyone read the same list.

High-RPM topics attract the most competition, the most established channels, and the most demanding viewers. A personal-finance video competes with channels that have been publishing for six years and have actual credentials. A video about historical shipwrecks competes with far less.

Worse, high-value niches usually cost more to make. A finance explainer that is wrong is a liability. Getting it right means research time, and research time is the cost that never shows up in anyone’s spreadsheet because it is your evenings rather than a subscription fee.

So the honest framing is: a high RPM is a discount on the views you need, paid for with a higher cost per video and harder competition. Sometimes that trade is worth it. It is not automatically worth it.

Start from the income you want, backwards

The useful first calculation is not “which niche pays most” but “how many views does my target income actually require, in each candidate niche”.

Work it backwards. Pick a monthly figure you would consider a success — not a fantasy figure, something that would make the effort feel repaid. Then use the channel income goal calculator to see what monthly view count that implies at different RPMs.

The output tends to be sobering, and that is the point. It usually reveals one of two things:

• The view count you need is achievable, in which case you have a real target and can plan a publishing rate against it.
• The view count you need is not plausible for a first channel in that niche, in which case you have learned something important before writing a single script rather than after forty videos.

Then run the same figure at a higher and lower RPM. The difference tells you how much the niche choice is actually worth to you in views — often less than people assume, because doubling the RPM halves a number that was already implausible.

Then subtract what the videos cost

The second variable is the one that quietly kills channels: production cost per video, in money and in hours.

Niches differ enormously here and it has nothing to do with their ad rate:

• Cheap to produce. Topics where the information carries the video and stills or simple b-roll suffice — history, explainers, case studies, “how X works”. A synthetic voice over images is genuinely adequate.
• Expensive to produce. Anything requiring generated video, heavy motion, or footage you cannot source freely. AI video generators bill per second of output and most generations get discarded, so budgeting several takes per usable clip is realistic.
• Expensive in hours regardless of money. Anything requiring accuracy — finance, health, law, anything where being wrong matters. The research is the cost.

Price your candidate niches properly with the AI video cost calculator before committing, and price a month rather than a video. A single video looks affordable in almost every stack. Three videos a week for a month is where the real number appears, and where the difference between niches becomes obvious.

The pairing that fails most often is a high-cost production stack in a low-RPM niche. The pairing that quietly works is a cheap stack in a middling niche, published consistently.

The variable nobody scores: will you still be doing this?

This is the one that decides most outcomes, and it does not fit in a table.

A faceless channel is not a passive asset. It is a publishing schedule. Whatever niche you pick, you are committing to research, write, narrate and edit videos about that subject twice a week, for months, largely without feedback, before anything happens.

So the question that matters more than RPM is: can you produce forty videos on this without resenting it?

Two practical tests:

• The forty-idea test. Sit down and list forty specific video titles in the niche. Not categories — titles. If you stall at fifteen, the niche is too narrow or you do not know it well enough yet. If you get to forty easily and want to keep going, that is the strongest signal available.
• The unpaid test. Would you read about this subject for an hour tonight if nobody paid you? If not, the research phase of every video will feel like work, and the videos will read like it. Viewers are very good at detecting a script written by someone who is bored.

This is not a motivational point. It is a practical one: the channels that succeed are almost always the ones that survived long enough for the recommendation system to find their audience, and survival is a function of whether the work is bearable.

How narrow is narrow enough

Too broad and nobody can describe your channel, so nobody subscribes — they watched one video and moved on. Too narrow and you run out of material in twenty videos.

The workable range is:

• Narrow enough that one sentence describes the channel and that sentence contains a specific subject, not a format. “Explainers about failed engineering projects” is a niche. “Educational content” is not.
• Broad enough to pass the forty-idea test comfortably.

A useful pattern is a narrow subject with a broad question. “Why did this business fail” applied to a whole industry gives you a recognisable channel and an effectively unlimited supply of episodes.

Where the tool fits

The faceless niche finder is built around this ordering rather than the RPM list. It asks what you can spend, how much time you have per video, and whether you are willing to narrate — because those constraints eliminate more niches than earnings potential does, and eliminating early is the point.

Use it to narrow the field, then do the arithmetic yourself:

1. Shortlist three niches you would genuinely watch.
2. For each, run your income target through the income goal calculator at that niche’s realistic RPM.
3. For each, price a month of production with the cost calculator.
4. For each, try the forty-idea test.
5. Pick the one that survives all four — not the one with the highest number in step two.

The niche that wins this process is rarely the top of anyone’s RPM list. It is usually something a little unfashionable that you happen to find interesting and can produce cheaply, which is exactly the profile of most channels that quietly work.