There’s a really good chance your most viewed videos aren’t your best videos. For a long time, creators (including me) default to whatever YouTube says is “performing” — views, impressions, watch time — and then we make more of that. The problem is YouTube is great at telling you what gets views, but it doesn’t tell you what happens after the view: leads, email signups, affiliate clicks, and the stuff that actually builds a business. In this video I break down the exact process I use now to decide what we make on Primal Video, and it comes down to five overlapping signals. First, I start with what I’m genuinely excited about (because that energy shows up on camera). Then I look at what’s already proven and should be updated or remade — especially anything older than a year that’s still driving leads. Next I use that data to generate related topic ideas (not random guesses), and I sanity-check everything through the viewer’s lens: pains, problems, and what they’d actually click. Finally, I validate with search data and keyword research so we’re showing up when people are actively looking for help — and then I turn the topic into stronger titles with AI. The big shift is using deeper tracking (like Video Stats) so we stop guessing and start making videos that grow the channel and the business.