You don't have a YouTube problem. You have an ideation problem

You can have an amazing hook, great editing and a beautiful thumbnail. But if nobody cares about the idea behind the video, you're packaging something nobody wanted in the first place.

The mistake I see is starting with the wrong question: “what should my next video be?” Then people brainstorm titles. Coming up with video ideas isn't about creativity or inspiration. It's about strategy. This is the process I use, and where the best ideas come from.

Step 1: build a keyword universe

Before I think about a single video, I want to understand the world of topics around my audience.

When people hear “keywords” they think SEO. That's not what this is. I'm trying to learn the language of the market. What are these people searching for? Which problems keep appearing? Which tools do they talk about? Which outcomes do they want?

AI makes this much faster. I give it five things: my niche, my ideal client, their desired outcome, their main problems, and what my business actually helps them accomplish. Then, instead of asking for “100 YouTube ideas,” I ask it to build a keyword universe.

That difference matters. I don't want AI deciding what my video should be. I want AI to tell me what to research. Whichever model you use, the results are similar.

I store the keywords in one place. I use Notion, where my whole content pipeline lives. Then I take one keyword at a time and look for signals: what keeps showing up, which questions people ask again and again, which videos are outperforming, what people are arguing about. For me, YouTube search gives the best results, but it depends on the niche.

Step 2: separate topics from packaging

Most people mix the topic and the packaging way too early. I keep three separate lists: topics, title formats and thumbnail formats.

• Topics. I use a YouTube account that only watches my niche, so the homepage shows me outliers: videos doing far better than the channel's normal. From each one I take only the topic. Not the title, not the thumbnail.
• Title formats. Here I go as far away from my niche as possible: finance, psychology, anything. To check if a title is a real format, I remove the keywords and search the structure. If “YouTube algorithm explained in 5 minutes” is an outlier, I search “explained in 5 minutes.” If other outliers use it too, it's a format, and it goes in the database.
• Thumbnail formats. Same method. I look for thumbnail structures that several outlier videos share.

Step 3: combine the three

Now the easy part. Take a proven topic, choose the title format that fits it, and choose a thumbnail format that fits both.

For example, I took the topic “hooks,” combined it with a title format I'd saved, and got: “The only video that will make you better at hooks.”

Two checks before you record anything. First, will this idea attract your ideal client? If not, it's not worth it, no matter how many views it could get. Second, make the title and thumbnail before you script. Show them to a couple of people. If the idea and the title are off, it's not worth your time to script, record and edit it.

Four places strategic ideas come from

The keyword universe tells you what the market cares about. These four sources make sure the ideas also bring clients.

1. Your competitors' best videos

Not copying them word for word. Find the top five channels in your niche, sort by most popular over the last six months, and look for patterns: which topics, which titles, which formats are working. Then adapt it to your angle. “How to get your first 1,000 subscribers” becomes “How to get your first 1,000 subscribers without posting every day.” The concept is already tested. You make it yours.

2. The steps of your own service

Break your main offer into the eight to ten steps your clients go through. For a done-for-you YouTube service: ideation, scripting, recording, editing, thumbnails, SEO, posting, analytics. Each step is a video, and each step is a pain point your ideal client is already searching for.

3. Proven title formats

Some formats work across hundreds of channels because they're clear, specific and promise a result:

• How to [result] in 2026
• How to go from [zero] to [result]
• Stop doing [the old way]. Do this instead
• Is [your niche] a scam?
• How [client] went from X to Y

4. What you already have

Most people are sitting on this without noticing. The questions in your DMs and emails: each one is a video. The objections on your sales calls: address them before the call. Every client transformation: a case study. If people keep asking “how long does it take to see results?”, that's a video. You already have the ideas in your inbox, your call recordings and your clients' wins.

Where it all leads: the free course video

Once you've made the individual videos, combine your whole process into one 60 to 90-minute free course. Someone who watches 90 minutes of you is pre-sold. By the time they book a call, they already know, like and trust you. At the end, give them the honest options: do nothing, do it yourself, or work with you.