Most creators come up with video ideas by asking themselves "what should I make next?" The fastest-growing channels work differently — they identify what YouTube's algorithm is actively amplifying, and make the best version of that.
This guide covers the five research methods that find genuinely high-potential video ideas — before you write a script or pick up a camera.
Method 1: The Outlier Research Method
Method 1 — Research-Based
Find Algorithm-Amplified Topics in Your Niche
An "outlier" is a video that got significantly more views than the average for its channel — typically 5–10× more. YouTube's algorithm pushed it to far more people than usual, which means it found something the algorithm rewards.
How to do it:
1. Open a competitor channel with a similar subscriber count to yours
2. Sort their videos by "Most popular"
3. Identify videos with 5–10× their average view count
4. Note the topic, title format, and thumbnail style — these are what worked
5. Repeat across 5–10 channels in your niche The pattern across multiple channels' outliers reveals what topics and formats are being amplified right now in your niche. These aren't random — the algorithm has a revealed preference for certain content.
How to do it:
1. Open a competitor channel with a similar subscriber count to yours
2. Sort their videos by "Most popular"
3. Identify videos with 5–10× their average view count
4. Note the topic, title format, and thumbnail style — these are what worked
5. Repeat across 5–10 channels in your niche The pattern across multiple channels' outliers reveals what topics and formats are being amplified right now in your niche. These aren't random — the algorithm has a revealed preference for certain content.
Method 2: Google Trends with YouTube Filter
Method 2 — Trend Data
Identify Rising Topics Before They Peak
Google Trends has a YouTube-specific filter that shows search demand on YouTube specifically (not Google Search, which has different user behaviour).
How to use it:
1. Go to trends.google.com
2. Type your topic and change the search type to "YouTube Search"
3. Check the related queries section — specifically "Rising" queries sorted by % increase
4. Topics marked "Breakout" have over 5,000% search increase — high risk, high reward if you move fast
5. Topics with steady upward trends (rising 50–200% over 90 days) are safer bets for long-term SEO value The goal: make a video while a topic is rising rather than after it peaks. A video published at 30% of peak search volume can rank for years. A video published after peak gets buried by competition.
How to use it:
1. Go to trends.google.com
2. Type your topic and change the search type to "YouTube Search"
3. Check the related queries section — specifically "Rising" queries sorted by % increase
4. Topics marked "Breakout" have over 5,000% search increase — high risk, high reward if you move fast
5. Topics with steady upward trends (rising 50–200% over 90 days) are safer bets for long-term SEO value The goal: make a video while a topic is rising rather than after it peaks. A video published at 30% of peak search volume can rank for years. A video published after peak gets buried by competition.
Method 3: Comment Mining
Method 3 — Audience Research
Find Ideas Your Audience Explicitly Asks For
The comments section of high-performing videos in your niche is a free research database. Viewers ask follow-up questions, express confusion about specific points, and request future content — all of which are validated video ideas.
Where to look:
— Your own comments: what do viewers ask about? What do they say they wish you'd covered?
— Competitor comments on their top videos: what are viewers asking for that the creator hasn't made yet?
— Reddit, Quora, and Facebook Groups in your niche: what questions come up repeatedly? A question asked 50+ times across comment sections in your niche is not opinion — it's a verified content gap. Fill it before anyone else does.
Where to look:
— Your own comments: what do viewers ask about? What do they say they wish you'd covered?
— Competitor comments on their top videos: what are viewers asking for that the creator hasn't made yet?
— Reddit, Quora, and Facebook Groups in your niche: what questions come up repeatedly? A question asked 50+ times across comment sections in your niche is not opinion — it's a verified content gap. Fill it before anyone else does.
Method 4: Your Own Analytics
Method 4 — Performance Data
Double Down on Your Own Outliers
Your own YouTube Studio analytics contain your most reliable idea generation data. Go to YouTube Studio → Analytics → Content → Sort by views (descending).
Your top 3 videos by views are candidates for:
— A sequel or follow-up ("What Happened After [Video Title]")
— A deeper dive into one aspect of the topic
— A contrarian or updated take
— A response to the most common comment or question on the original Your audience already told you what they found valuable. The most predictable growth strategy is making more of what already worked — not constantly experimenting with new topics.
Your top 3 videos by views are candidates for:
— A sequel or follow-up ("What Happened After [Video Title]")
— A deeper dive into one aspect of the topic
— A contrarian or updated take
— A response to the most common comment or question on the original Your audience already told you what they found valuable. The most predictable growth strategy is making more of what already worked — not constantly experimenting with new topics.
Method 5: Cross-Platform Idea Validation
Method 5 — Validation
Check If the Topic Has Demand Before Filming
Before committing to any idea, run a 5-minute validation:
1. YouTube autocomplete (incognito) — does this topic have suggestions? High volume = confirmed search demand
2. Top results quality check — are the existing top videos old or low-quality? Can you make something better?
3. Comment sentiment on existing videos — are viewers satisfied, or are they asking for more? A topic with autocomplete suggestions + weak existing results + unsatisfied viewers in comments is the highest-confidence video idea you can act on.
1. YouTube autocomplete (incognito) — does this topic have suggestions? High volume = confirmed search demand
2. Top results quality check — are the existing top videos old or low-quality? Can you make something better?
3. Comment sentiment on existing videos — are viewers satisfied, or are they asking for more? A topic with autocomplete suggestions + weak existing results + unsatisfied viewers in comments is the highest-confidence video idea you can act on.
Idea Quality vs. Idea Quantity
Volume isn't the bottleneck for most creators — idea quality is. Ten video ideas that are variations of what's already in your niche won't help as much as one outlier-researched idea with a fresh angle on a proven topic.
| Idea Source | Reliability | Speed | Best For |
|---|---|---|---|
| Outlier research | High (algorithm-validated) | 30–60 min/week | Browse/recommendation growth |
| Google Trends (YouTube) | Medium-High (demand confirmed) | 10–15 min/week | Search traffic growth |
| Comment mining | High (audience-requested) | 20–30 min/week | Community retention and loyalty |
| Own analytics | Very High (proven format) | 5 min | Predictable growth on existing audience |
| AI brainstorming | Medium (not validated) | 5 min | Idea generation starting point — still needs validation |
The weekly idea research habit: Thirty minutes per week of structured outlier research and Google Trends scanning generates more high-quality ideas than hours of unstructured brainstorming. Block 30 minutes every Monday. Check 3–5 competitor channels for outliers, scan Google Trends for rising topics, and read comments on your top 2 recent videos. Your content calendar fills itself.
Frequently Asked Questions
What makes a YouTube video go viral in 2026?
Algorithm amplification driven by exceptional CTR (thumbnail + title that almost everyone clicks when shown) combined with average view duration well above the niche average. Most viral videos are planned — they succeed because of strategic topic selection and high-quality execution, not luck.
How do I find what people are searching for on YouTube?
Three free methods: YouTube autocomplete in incognito (suggestions are ordered by search volume), YouTube Studio's search terms report (shows exactly what people searched to find your existing videos), and Google Trends with the search type set to "YouTube Search" to find rising topics in your niche.
What is the outlier research method for YouTube?
It identifies videos in your niche that got 5–10× more views than average for the channel that published them. These are algorithm-amplified topics — videos the recommendation system actively pushed to large audiences. By studying what topics, title formats, and thumbnails triggered outlier performance, you can make informed decisions about what to make next.
Should I copy successful YouTube videos in my niche?
Study them — don't copy them. The correct approach is to identify what made the outlier work (format, emotional hook, topic angle) and apply those learnings to genuinely original content in your voice. A direct copy underperforms the original because the audience has already seen it.
Outlier Research Built Into Your Browser
Teka Creator Tools shows you which videos in your niche are over-performing their channel average — outlier research in seconds, without manual spreadsheets. Join early access.
Join Early Access