How to Find YouTube Video Ideas With the Outlier Method (Worked Example)

Most creators pick video ideas by asking, "what do I feel like making?" That is a coin flip. The outlier method replaces the coin flip with evidence, and it does not require a paid tool to start.

Here is the idea in one sentence: when a video dramatically outperforms its own channel's baseline, the topic — not the creator's size — is doing the heavy lifting.

Why raw view counts lie

A 500,000-view video looks impressive. But if it came from a channel averaging 800,000 views per upload, it actually underperformed. Meanwhile, a 50,000-view video on a channel averaging 3,000 is a 16x hit. Same number range, opposite meaning.

Raw views blend two things together: the creator's existing reach and the topic's inherent demand. An outlier separates them. It says: this specific topic reached far beyond the people who already followed the channel. That is exactly the kind of topic you want to make.

A worked example

Say you run a channel about home cooking. You have 400 subscribers. Your videos average 150 views. Let's find topics that work.

Step 1: Build a target list. Pick five to ten cooking channels slightly ahead of you — say 2,000 to 20,000 subscribers. Big enough to have data, small enough to be fighting the same algorithmic battles. Ignore the mega channels; their results are not comparable to yours.

Step 2: Establish each channel's baseline. Open one target channel's video tab and look at its last ten uploads. Note the rough median view count. Do not average in the one viral video — the median is the "normal" performance you are measuring against.

For our example channel, recent uploads sit around 4,000 views. That is the baseline.

Step 3: Scan for gaps. Now look for videos that beat 4,000 by a wide margin in the last few months:

• "3 Easy Weeknight Dinners" — 3,200 views. On baseline.
• "How to Meal Prep for One Person" — 2,800 views. On baseline.
• "The 15-Minute Rule That Fixed My Grocery Bill" — 61,000 views. A 15x outlier.
• "My Favorite Kitchen Knife" — 5,100 views. Mild.

One topic punched 15 times above the channel's normal reach. That is your signal.

Step 4: Ask why it worked. An outlier is a clue, not an answer. Read the packaging and the first minute to figure out what the shared audience reacted to. In our example, the winner has three ingredients the others lack:

• A specific, numeric promise ("15-minute rule") instead of a vague one.
• A pain point the viewer already feels (grocery bill) rather than a neutral activity (meal prep).
• A system the viewer can apply today, not inspiration.

The topic was not "cooking." It was "save money on groceries with one rule." That distinction is everything.

Step 5: Make a better, original version. You do not copy the video. You take the pattern — a specific numerical rule aimed at a felt pain point — and apply it to your own angle: "The 15-Minute Rule That Cut My Grocery Bill" becomes "I Tried the 15-Minute Grocery Rule for 30 Days." New footage, new perspective, proven topic shape.

How to read an outlier correctly

Not every gap means the same thing. Before you act, sort the outlier into one of three buckets:

The topic outlier. The subject itself is in demand. ("How to meal prep for one person" suddenly trending.) This is repeatable — you can cover the same territory from a new angle.

The packaging outlier. The topic is ordinary but the title and thumbnail made it irresistible. This is also repeatable, but your fix is in the packaging, not the subject.

The event outlier. The video rode a trend, a release, or a news moment. This is not repeatable, and chasing it is how creators waste a month. If the spike lasted 48 hours and then stopped, it was an event.

Only one of these three is worth building a strategy on in any given case, so name it before you film. If you want to see the full five-step teardown this feeds into, we walked through a complete competitor analysis example.

The rule of three

For a small channel, use three times the channel's baseline as your minimum bar for "worth studying," and treat ten times as a strong demand signal. Below three times, you are probably looking at normal variance, not a pattern.

Also compare like with like. A Short exploding is not evidence that a long-form video on that topic will work, and vice versa. Keep Shorts and long-form data separate.

Turn it into a repeatable system

Do this once a month and you will never sit in front of a blank page again:

1. Maintain a list of 5 to 10 peer channels.
2. Record each channel's median views.
3. Each month, flag anything over 3x that baseline.
4. For each flag, write down the topic, the packaging pattern, and whether it was a topic, packaging, or event outlier.
5. Make the ones that fit your lane and your strengths.

The point is not to copy anyone. It is to stop guessing which ideas have an audience. When you build your next video on a topic a shared audience has already proven they will click and watch, you are playing with the odds in your favor instead of against them.

---

Edison runs this process continuously — scouting your niche, flagging outliers the moment they appear, and handing you the topic and packaging pattern behind each one so you know exactly what to make next.