By Adam Ivy. I have been on YouTube since 2008, spent eight years doing it wrong, then grew past 300,000 subscribers and generated over 3.5 million dollars from my content. Everything below comes from that scar tissue.
Before you film anything, run the idea through four questions: Is it for a specific person? Is there real demand? Would a stranger stop scrolling for the title and thumbnail? And what unique value are you adding that existing videos are not? If an idea fails the filter, do not film it. Every video upload is a tryout with YouTube, and validation is how you stop wasting tryouts.
One massive mistake I see people make time and time again is rushing into the production side of things without a plan, without a roadmap. I did it for years. Here is the filter that fixed it.
Who exactly is this video for?
A specific person, never "everybody." Not "anyone interested in fitness" but "a new entrepreneur starting a business" or "a dad with 30 minutes a day trying to lose 20 pounds." If you cannot name the person, YouTube cannot find them either, because the system works by matching your video to a specific viewer profile. Confusion kills reach faster than bad quality ever could.
A quick gut check I use on the channel level: if you cannot explain the value of your channel in five words, your channel is a hot mess, and every individual video inherits that confusion.
Is there real demand for this topic?
Check before you assume. Look at search volume, look at what is trending in your niche, and look at whether comparable videos from channels your size actually get watched. Passion alone is not a strategy. A video can be deeply important to you and invisible to demand, and the tryout does not grade on sincerity.
This does not mean you only chase trends. It means you confirm that the specific person from question one is actively looking for this answer.
Would a stranger stop scrolling for this?
That is the scroll-test, and it has to pass before you film, not after. Draft the title and thumbnail concept first. If a stranger who has never heard of you would not stop scrolling, the idea is not ready, no matter how good the content plan is. Your title and thumbnail are the front door, and this question is how you check the door before building the house.
In practice I generate far more title and thumbnail combinations than I ever film, then only produce the strongest. Planning a hundred combinations to film the best ten sounds excessive until you compare it with spending a week producing a video nobody clicks.
What unique value are you bringing?
Something the existing videos on this topic do not have. If your video would be redundant, skip it or find the angle only you can take: your results, your numbers, your story, your contrarian read on the common advice. Every random me-too video you post is training your audience to ignore you.
What does validation look like in practice?
Here is the filter as a working checklist:
| Question | Passes when | Fails when |
|---|---|---|
| Specific person | You can name the exact viewer | It is "for everybody" |
| Real demand | Search and niche data confirm interest | You assume people should care |
| Stranger scroll-test | Title and thumbnail stop a cold viewer | Only your subscribers would click |
| Unique value | You add proof, story, or a new angle | The video already exists ten times |
Massive imperfect action beats perfect research every single time, but that only applies to execution. Ideas are cheap to kill and expensive to film. Kill the weak ones on paper.
Still have questions?
How long should validation take? Minutes, not days. The filter is four questions, not a research project. If an idea needs an hour of convincing, it already failed.
What if an idea fails one question but I love it? Fix the failing question instead of ignoring it. Usually the idea is not wrong, the framing is. Narrow the person, sharpen the angle, or repackage until the scroll-test passes.
Can one topic really produce enough validated videos? Yes. Ten core topics multiplied by different angles, such as solving a problem, taking a stance, showing proof, and telling your story, gives you forty distinct videos before you ever run dry. Depth beats novelty.