A niche is not a permanent identity you pick on day one. It is the narrowest lane you can test fast, then widen once one topic keeps outperforming the rest. Most new creators freeze at this step and never publish.
I froze for two months on a channel name I eventually changed anyway. The naming was not the bottleneck. Publishing was.
You cannot think your way to a niche. You can only test your way to one.
'Productivity' is not a niche. 'Task management for freelancers who juggle five clients' is. The narrower version tells you what to film next Tuesday and who to send it to.
Narrow also means fewer competitors fighting for the same search terms. A video titled for freelancers can rank where a generic productivity video gets buried under channels with a million subscribers.
Pick three topics inside your general interest. Make four videos per topic across two weeks. Do not judge by views alone. Watch which topic produced comments, DMs, and saves, because those are the signals a small channel can actually generate.
At the end, one lane will have stranger engagement that the others lack. That is your niche for the next six months, not forever.
Once a lane works, adjacent topics are safe. If freelancer task management works, project templates and invoicing tools are close enough that the same viewers care.
Widen one step at a time. Jumping from freelancer tools to general productivity drops the reason your early viewers stayed, and early viewers are the ones who tell the platform what your channel is about.
I stopped naming channels after subjects and started naming them after people. A channel called 'for freelancers who hate admin' attracts a clearer viewer than one named after software.
The test is whether a stranger can guess who the channel is for in three seconds. If they cannot, you have described a topic, not a niche.
A series gives viewers a reason to come back and gives the recommendation system a reason to keep suggesting you. Here is how to design one that actually grows.
Your first 100 subscribers come from people who already trust you, not the algorithm. A concrete plan for earning each one without gimmicks or paid ads.
Collaboration moves a new channel faster than posting alone. Find peers at your level, trade a real favor, and borrow an audience that trusts them.
Analyze a channel, extract its structural DNA, and generate scripts engineered from patterns that actually work.
Analyze a Channel