How to Pick a Niche for a Faceless YouTube Channel (Real RPM Data + Research Workflow)

Niche selection is the entire game. The agent produces a great video on any topic. The algorithm only pays well in some of them. Pick wrong, and you have a channel running on autopilot toward zero. Pick right, and you have an asset that compounds.

This piece is a starting map, not a closed catalog. We walk through 9 niches we have seen working on faceless long-form in 2026, with revenue numbers from a small set of public channels we audited, the research workflow operators actually run (VidIQ is the workhorse), and the niches we recommend avoiding. Faceless YouTube has dozens more working niches and thousands of channels we did not sample. Use this to start, then run the workflow below to find your own.

How niche economics actually work

Two numbers decide what a faceless channel earns:

• RPM (revenue per 1,000 views). Set by your audience and niche.
• Volume (total views). Driven by retention, cadence, and thumbnails. Set by your format.

A channel can win with high RPM and modest volume. A channel can also win with low RPM and viral volume. The mistake is picking a niche without checking which lever you are pulling.

A few public examples to anchor the math:

• Crayon Capital (finance): ~$10 RPM, ~$34,000 per month, 18 mature long-form videos. High RPM, modest volume.
• Meowia (animals): ~$2 RPM, ~$47,000 per month, 4 mature videos. Low RPM, viral volume.
• xkcd's What If? (science): ~$4 RPM, ~$24,000 per month, $264 revenue per video-minute. Long retention, sticky topic.
• Species | Documenting AGI (true crime): ~$5 RPM, ~$29,000 per month. Solid mid-RPM with reliable volume.

Same agent-style production. Very different earning profiles.

9 niches we have seen working on faceless long-form

Listed by typical RPM, with public 2026 revenue estimates from a small audit of established channels. There are more working niches than this; treat the list as a starting set, not a complete one.

1. Finance and business explainer ($10 to $25 RPM)

The highest-paying corner of faceless YouTube. Money topics, investing, real estate, credit cards, business explainers, founder stories, "how rich people got rich" content.

• Crayon Capital: ~$34K/mo
• Legend Investor Mindset: ~$25K/mo

Why it pays: a single credit-card customer is worth $500 to $2,000 LTV to advertisers, so banks and fintechs bid $30 to $50 CPM. Audience is US-skewed, financially active, 30 to 60.

Caveat: avoid generic "10 stocks to buy" template content. YouTube's January 2026 sensitive-topics monetization update specifically targets template-driven faceless finance. Add original analysis, sourced data, and disclaimers.

2. Auto and transport ($5 RPM)

Cars, trucks, engines, market commentary, classic-car deep dives.

• Truck Guide: ~$13K/mo

3. Crime and dark documentary ($5 RPM)

Cold cases, financial crime, white-collar crime, cult deep dives, conspiracies.

• Species | Documenting AGI: ~$29K/mo
• Professor POV: ~$15K/mo

4. Psychology and education ($5 RPM)

Behavior, habits, self-improvement, learning.

• easy, actually: ~$11K/mo
• PsychologyAboutYou: ~$6K/mo

5. History documentary ($4.50 RPM)

Wars, founders, lost civilizations, scandals.

• Majestic Studios: ~$13K/mo
• StickTory: ~$7.7K/mo
• Zack D. Films: ~$6.5K/mo

6. Science and explainer ($4 RPM)

Space, biology, physics, geology, why-things-work.

• xkcd's What If?: ~$24K/mo
• Mr. Science: ~$24K/mo
• Spinosnack: ~$11K/mo
• Cosmicus: ~$10K/mo

7. Survival and building documentary ($4 RPM)

Shelter, off-grid, extreme environments, immersive build.

• BikeGen Mountain: ~$25K/mo

8. Religion and sleep prayer ($3 RPM)

Long-form passive listening, overnight content. Lower RPM but very low production cost per viewer.

9. Animals and pets ($2 RPM)

Animal stories, pet behavior, rescue content. Low RPM, but viral volume can dominate the leaderboard.

• Meowia: ~$47K/mo (the absolute revenue leader on this list, despite the $2 RPM)
• Felune: ~$11K/mo

Niches to avoid

Niches with a low income floor:

• Kids content. RPM near zero since the FTC ad-rule update in 2020.
• Music compilations. Copyright takedowns plus low retention.
• Gaming aimed at kids. Same demographic problem.
• Reaction or trending-news commentary. Burns out per cycle.
• Generic motivation. Saturated, low CPMs, no defensible audience.
• Shorts-heavy content in any niche. Shorts pay 50 to 100x less than long-form.

You can build a channel in any of these. You will not build a high-RPM asset. Different game.

The niche research workflow

This is what operators actually run, in order, before committing to a niche.

Step 1: Confirm long-term demand

Open Google Trends. Search the seed topic over 5 years. Reject anything in clear decline.

Step 2: YouTube search the seed

Look at the top 20 search results for the niche keyword. Count: how many are faceless, how many sub-100K-sub channels broke 100K views on a recent upload, and how long are the videos.

Step 3: Pull a competitor set

Use ChannelCrawler to filter 22M-plus channels by country (US for high RPM), sub band (1,000 to 100,000 captures rising operators), and category. Export 30 to 50 channels.

Step 4: Cadence and velocity check

Plug each into Social Blade. Look for channels gaining 2,000-plus subs per day with 2 to 4 uploads per week. Anything below 2 uploads per week with flat sub growth is a dead niche.

Step 5: Outlier hunt

Use 1of10 or VidIQ Outliers to find videos that performed 10x their channel average. If small channels (under 10K subs) are producing recent outliers, the niche is still penetrable. If only mega-channels win, the barrier is too high.

Step 6: Keyword validation

Use VidIQ or TubeBuddy to check keyword volume and trend direction. Aim for 1,000 to 10,000 monthly searches with a score above 40, trending up or stable. TubeBuddy shows trend direction; VidIQ does not.

Step 7: Format audit

Look at thumbnail consistency, hook style, average view duration. Flag heavy Shorts mix as RPM-poison.

Step 8: Score the niche

Rate 1 to 5 on demand, repeatability, defensibility, monetization. Anything below 16 of 20 needs rework.

Step 9: Cross-check exit value

Empire Flippers public sales data shows finance, real estate, and B2B faceless channels selling at 30 to 40 times monthly profit. Entertainment sells at 12 to 18 times. If the niche cannot resell well, the asset framing breaks.

The tools that matter

VidIQ is the workhorse. Most faceless operators run it as the daily driver: keyword volume, search competitiveness, channel monitoring, recent-video outliers, niche scoring. The free tier is enough to start. The paid tier ($10 to $39 per month) is where serious operators live.

Add the rest of the stack as needed:

• TubeBuddy — keyword research with trend direction (the one signal VidIQ does not surface). $4 to $30 per month.
• Social Blade — sub growth, daily view delta, estimated earnings sanity-check. Free + paid tiers.
• ChannelCrawler — filter 22M-plus channels by country, niche, sub band. Free + paid tiers. Useful when you need to pull a competitor list cold.
• 1of10 — outlier video discovery beyond what VidIQ surfaces (videos performing 10x their channel average). $39 per month.
• OutlierKit, TubeLab, NexLev — niche finders with RPM estimates and micro-niche scoring. $15 to $49 per month each.
• Tubular — cross-platform analytics + audience demos. Enterprise-tier, agency-grade.
• Google Trends — free, multi-year demand direction. Run the seed topic through this before anything else.
• YouTube search itself — free top-20 audit. Always step one.
• Quickads.ai — freemium AI niche brainstorm if you are stuck.

A solo operator does not need all of these. VidIQ plus Social Blade plus ChannelCrawler plus Google Trends covers most of the workflow. Add 1of10 once you have a channel running and need outlier inspiration.

Public revenue estimates from Social Blade and similar tools can be off by 10x or more. Use them for direction, not numbers.

Niche red flags

• Converging thumbnails and titles in the top 20 — saturation past angle level.
• Only mega-channels (over 1M subs) get views. No recent outliers from sub-10K channels means the barrier is too high.
• Flat 5-year Google Trends curve plus declining sub growth across top channels = decaying niche.
• Heavy Shorts mix in a high-CPM niche — RPM will collapse.
• Template TTS plus stock B-roll formats. YouTube flagged these as "inauthentic content" in July 2025.
• Shock, tragedy, or sensitive content. January 2026 monetization changes cap revenue.
• High-CPM niches dominated by credentialed channels (legal, medical, insurance) — RPM is real but you cannot compete without subject-matter authority.
• Generic motivation, Reddit narration, AI slideshow compilations. All called out as oversaturated.

Picking your first niche

Start with one. Pick the bucket where you have the strongest hunch on story angles, then narrow to a specific angle.

Example narrowings:

• Finance: "Forgotten financial scandals that built modern Wall Street"
• Crime: "White-collar finance crime since 1980"
• History: "The richest people in history that nobody remembers"
• Survival: "Off-grid shelter builds in extreme cold climates"
• Science: "How ancient infrastructure actually worked"
• Animals: "Rescue stories you have never heard of"

Each of those is a 200-video backlog with a defensible angle.

Once your first channel runs for 90 days and produces signal, you spin up the second channel in a different bucket. Then a tenth. Then thirty.

Pick the niche. Own the asset. The AI does the work.

Plans from $149 per month at noodletomato.com.