Why Your YouTube Titles Are Failing — Jake Thomas Explains

Jake Thomas is not a YouTube growth guru who stumbled into titles because they sounded trendy. He’s the founder of Creator Hooks, a newsletter that breaks down the psychology behind why videos get clicked — and he learned most of what he knows by being publicly terrible at it first.

Before Creator Hooks, Jake was a marketing manager at a YouTube channel in the fishing niche. His boss published twice a day and roasted every single title Jake submitted. Twice a day, every day, until Jake stopped guessing and started studying. What came out of that grind — and a subsequent run A/B testing titles for a channel with 10 million subscribers — is one of the more honest and practical frameworks for title writing available in the creator space right now.

This is not a post about power words or headline hacks. It’s about what Jake actually figured out across 2,000-plus titles, a 900,000-video database, and a psychology degree he somehow made useful on the internet.

The Reason Your Titles Don’t Work Has Nothing to Do With Creativity

Most creators approach title writing as a creative exercise. They sit down with a finished video and try to invent something clever. Jake’s argument — backed by years of watching this fail — is that this is exactly backwards.

“In an ideal world, your title is written before you actually make the video. So that you have a great title that grabs your audience’s attention, that wants them to click, then you have an intro that confirms that idea, and then you have content that delivers on that idea. Otherwise, that’s how you get into clickbait — you’re trying to put lipstick on a pig.”

Writing the title after the video is done is the structural reason so many creators drift into clickbait without meaning to. The title becomes a sales pitch for content that wasn’t built around a clear promise. The gap between what the title implies and what the video delivers is where audience trust breaks down.

Jake’s fix is mechanical: know your title before you shoot. That forces your intro, your content structure, and your delivery to actually align with what made someone click in the first place.

The Dream Ten and Model Ten: A Repeatable System for Video Ideas

One of the most practical frameworks Jake shared is what he calls the Dream Ten and Model Ten — two lists of channels every creator should build before they publish their first video on a new channel.

The Dream Ten is your competitor list. These are the channels talking about the same topics you cover. This is where you pull proven topics — things your audience already cares about and actively searches for.

The Model Ten is a list of channels outside your niche whose content structure, title style, and packaging you admire. For Jake’s own YouTube education channel, his Model Ten includes finance channels, fitness channels, marketing channels, and even a plants channel. Adjacent, not identical.

The system works like this: take a proven topic from your Dream Ten, find a proven formula from your Model Ten, and remix them. The topic is familiar to your audience; the format is borrowed from somewhere they’ve never seen it applied. That combination is what produces titles that feel both relevant and fresh.

The deeper use of these two lists goes well beyond titles. Jake describes using them as a surrogate mentorship system:

“Pretty much any question you might have for growing your channel, you can pretend that those people are your mentors and just ask yourself — let’s say Mr. Beast was on your model ten. You might say, ‘Hey Mr. Beast, how long should my intros be?’ And then you go look at his channel and you see how long his intros are and you can pretty much guess what he would tell you.”

Title length, thumbnail composition, whether to show your face, what sponsorships to pursue, how long your intros should run — all of it is answerable by studying these two lists with intention.

Why Modeling Is Not the Same as Copying

Jake is direct about originality: stop chasing it, especially early. His analogy is learning guitar. You don’t start by ripping solos. You play Happy Birthday. You play Mary Had a Little Lamb. You figure out what sounds good to other people before you have any business trying to invent your own sound.

The distinction he draws is between copying within your niche and modeling from outside it. Copying inside your niche — taking another fishing channel’s title and running it on your fishing channel — is both lazy and visible. Modeling from outside means taking a proven formula from a laptop review channel and applying it to a fishing lure video. Your audience will never trace it back. The psychology transfers; the surface content is entirely yours.

He also makes a point worth sitting with for creators worried about being derivative: if you have 50 subscribers, nobody is watching your videos closely enough to accuse you of copying anything. You need reps, not originality. The reps are what eventually teach you why something works — which is when you can start making real creative decisions.

Curiosity, Fear, and Desire: How to Actually Use Them

Jake has studied titles at scale — his current database sits at roughly 900,000 videos — and his conclusion is that almost every high-performing title is pulling on at least one of three emotions: curiosity, fear, or desire. Not all three at once. Usually one, executed well.

Fear, handled wrong, becomes the kind of clickbait that makes people feel manipulated after they click. Jake’s approach to using fear without being manipulative comes down to specificity and framing. The most effective version is to name a pain point your audience already has:

• Pain point framing: “If nobody’s clicking your videos, watch this.” The viewer doesn’t feel accused. They feel seen.
• Mistake and regret framing: Titles built around mistakes or regrets tap into fear of failure without being alarmist.
• First-person softening: “My biggest regret when I started my first YouTube channel” lands very differently than “Your biggest regret was starting a YouTube channel.” Same emotion, completely different reception. One feels like a personal story; the other feels like an attack.

Desire-based titles require more precision. Jake’s point is that “how to grow your YouTube channel” is technically about desire but it’s too vague to trigger a real emotional response. “How to get your first 100,000-view video” speaks to a specific want. “How to make $5,000 a month from coaching calls on YouTube” is even more specific. The more precisely you name the outcome, the more directly you’re speaking to someone who actually wants that outcome — and the more likely they are to click.

The word “secret” is a case study in overuse. Jake’s 900,000-video analysis showed it underperforming relative to other words that serve the same psychological function — curiosity, the promise of insider knowledge. His read: Russell Brunson made “secrets” famous as a marketing device, everyone adopted it, and now it’s so saturated that audiences have developed mild immunity to it. It can still work, but it’s no longer a reliable lever.

Why the Words You Choose Matter More Than the Formula

Every word in a YouTube title carries emotional connotation. “The secret to getting more views,” “how to get more views,” and “I wish I knew this before starting my channel” are all functionally trying to do the same thing — but they feel completely different to the person scanning their feed. Jake’s database work was specifically aimed at figuring out which words, at scale, are actually associated with above-average performance.

Words like “never,” “stop,” “every,” and “any” consistently showed up in stronger-performing titles. They’re also useful as idea prompts — starting with a high-performing word and working backwards to a video concept is a legitimate approach Jake recommends.

On the question of dated language: Jake is pro-year-tagging for videos in fast-moving niches (tech, software, social platforms), even though it’s the opposite of evergreen. His reasoning is that the performance boost is large enough to justify remaking the same five to ten videos every December through February. For genuinely timeless topics, skip the year. For anything that changes — put the year in and plan to refresh it annually.

Educational Channels vs. Entertainment Channels: A Different Game Entirely

One of the more clarifying distinctions Jake makes is between channels where the audience shows up for the creator versus channels where the audience shows up for the topic.

Emma Chamberlain and MrBeast operate in a world where the title is almost irrelevant. Their audience is there for them. It doesn’t matter what they title the video — people are going to watch because they want to spend time with that person.

Educational channels work on completely different logic. If you run a fitness channel and you make a video about your morning routine instead of about squats, and your audience is there for squats, that video is going to underperform — and it has nothing to do with your title. It’s the wrong topic for the audience you’ve built.

The reason this matters for title strategy: educational creators often see wide variance in their view counts and blame their titles when the real issue is topic selection. Understanding why your audience watches — what goal they’re trying to achieve, what problem they’re trying to solve — is upstream of every title decision.

For podcasts specifically, Jake’s advice is to lead with the insight, not the guest’s name, unless that guest is a household name. If you’re interviewing someone with real expertise but no mass recognition, ask yourself what your audience will actually get out of the conversation and build your title around that. And ideally, figure out your five candidate titles before the interview even starts — then ask questions designed to deliver on whichever title you choose.

What 2,000 Titles and a Psychology Degree Actually Teach You

The through-line in Jake’s work is that psychology doesn’t change. He references a Paul passage from the Bible where the complaints are essentially the same complaints people have today, 2,000 years later. Copywriting books from 1960 and copywriting books from 2020 both describe audiences overwhelmed by too many ads. Human attention, fear, desire, and curiosity operate on biology that isn’t updating on a quarterly release cycle.

That’s his core argument for why learning title psychology is worth the investment — not just for YouTube, but for email subject lines, LinkedIn posts, Twitter threads, and SEO. He’s used the same principles to rank number one on Google. The platform changes; the psychology doesn’t.

His advice for anyone trying to get better at this is also notably unglamorous: get your reps in. He wrote approximately 2,000 titles over three years getting roasted by his boss. Then he A/B tested title variations for a 10-million-subscriber channel. Then he built a database of 900,000 videos. The pattern recognition he has now is not the result of a framework he stumbled onto — it’s the result of volume and feedback.

He’s also started publishing his own YouTube channel and describes the experience of getting his own titles critiqued by a small accountability group of peers as one of the most valuable things he’s done. His takeaway: the creator who feels mentally “done” with an idea is often the worst judge of whether that idea is actually good. External feedback from people at a similar stage, working on similar content, consistently surfaces improvements the creator can’t see.

The practical infrastructure he recommends for anyone serious about improving: build your Dream Ten and Model Ten lists, subscribe to the Creator Hooks newsletter (free, weekly, breaks down what worked and why), and find two or three people working on similar content who will tell you honestly when your title is bad.

None of that is complicated. All of it requires actually doing it, over and over, until the pattern recognition develops. That’s the part most creators skip.