Depliny (@deplinyy) header
Depliny (@deplinyy) profile picture

Youtube strategist & thumbnail designer | Building @blazifyyt - generated 70M+ views, $1M+ in sales | Work with me👇

Full time
Joined May 2023
Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)Photo by Depliny (@deplinyy)
that's the real power of trending niche this channel was launched 2 months ago
4
13
211
45,420
claude's advices to youtubers for the gta 6 launch
stupid youtuber on gta 6 launch day: plays the whole game first, spends a week editing, uploads after every gaming channel has already covered it, then wonders why nobody watched. smart youtuber on gta 6 launch day:
2
12
191
24,792
bold prediction: if Iman Gadzhi and Jake Paul actually team up, the 18-25 biz op space won’t just get disrupted, it’ll get owned. Nobody else will come close.
4
13
2,057
idk if my vidiq is broken, but right now an 8-year-old PewDiePie video is getting more views per hour than his goodbye video
2
6
964
some results for our clients for last year in 1 tweet
10
1
45
2,192
stupid youtuber on gta 6 launch day: plays the whole game first, spends a week editing, uploads after every gaming channel has already covered it, then wonders why nobody watched. smart youtuber on gta 6 launch day:
3
5
2
71
49,812
goes live the minute it drops. then the smart youtuber keeps going: the next morning he cuts the best moments into shorts. he uploads with a title in the format people are already searching for. and his video gets the spot next to every other gta 6 video people are watching, while the stupid one's disappears under everyone else's.
1
1
14
4,530
on launch week, speed beats polish. an evergreen video can take weeks to perfect. A trending one can't. The topic has a window, and the window closes. so the question isn't how good your video is. It's how fast you can get it out. what type will you be when GTA 6 launches?
8
3,975
mrbeast didn't need a new idea. he took "last to leave grocery store" (207m views in 6 months), raised the prize to $1,000,000, and reworked it into "eat everything in a grocery store". In this way, he improved the original idea 342m views in one month. the idea you need might already be on your channel.
1
1
15
976
hot take: consistency without testing is just practice at being average. post less, test more.
1
251
ok, can fable 5.2 repeat this? guess which thumbnail performed the best
15
73
1,693
nearly every celebrity flops on youtube. well-known name might earn a click, but people only come back if they loved the video. that's why neymar has 241m on instagram and only 9.3m on youtube. mrbeast has 526m on youtube alone, more than neymar has on instagram, x and youtube combined. same story with almost other all celebrities or brand, youtube is a completely different game a name gets the click. A video with a strong idea from a creator who consistently produces content and builds an audience.
I have a hypothesis for why big brands and celebrities keep losing on youtube to creators. Ray-ban's glasses video: 1.8m views. Marques Brownlee on ray-ban glasses: 8.5m. Ronaldo's own channel: 34m. Mark Rober with ronaldo: 51m. it's not that creators have no team. They have big content teams too. My guess is the difference is who's in the room. The creator is inside the process, and usually nobody tells them what they can or can't say. youtube is a testing game. A creator can test the craziest idea, say too much, show something wild, sound like a person and not a press office. A company has rules, approvals and a brand to protect, so it tests the safe version. same goes for celebrities. Sponsors, contracts, a public image to protect. Every video has to follow someone's rules, so the safe version wins there too. safe wins inside a company or a contract. It loses in a feed where only the real ideas get seen. am I wrong? name a big brand that gets youtube right.
1
2
302
friendly reminder: repeating a format that has already worked is the biggest hack in the big 2026, just look at this guy
15
1,436
Depliny retweeted
who’s the biggest YouTuber with your birth year? the most subscribed creators from 1995 to 2014:
2
1
7
635
Depliny retweeted
youtube advices I wish I'd known when I started: 1. packaging matters more than any other part of video. People don't watch your video, they buy the click first. 2. don't invent a new format, find what's already working in your niche and do it better. 3. create every video for one specific audience, not for everyone. 4. your title and thumbnail should work in combination with your video intro. 5. your thumbnail should have no more than 3 main elements. 6. re-hook the viewer every 30-40 seconds, attention is always looking for a reason to leave. 7. one video, one problem, one solution. don't cram three ideas into a single upload. 8. if retention drops at the same spot across multiple videos, that's not a coincidence, that's a pattern worth fixing. 9. subscriber count is a vanity metric, average views per video tells you the real health of a channel. 10. never settle on your first thumbnail concept, make at least 2-3 and test them. 11. if your ctr is below your channel average, fix the thumbnail before touching anything else. 12. a flop with high watch time isn't a bad idea, it's bad packaging. 13. short form should funnel into long form, not exist as its own separate thing. 14. hire a expert for each part of the video or a team sooner than you think you're ready to. 15. post consistently and fix one thing at a time, don't wait to make it perfect before you publish.
4
7
63
3,428
Depliny retweeted
a/b test for the “Vaping was supposed to help. What went wrong?” video which thumbnail would you click on?
21
1
83
1,577
Depliny retweeted
mrbeast quietly launched a channel and achieved 10m subscribers with just 4 long videos. it's his animation channel, Beast Animations. The 7th one he's taken past 10 million. here's what seems to be going on: it's an animated series on youtube, and mrbeast is the main character. so why does it work? my opinion: → every video is an event "I spent $7,000,000 on my animated series." 17m views in 6 days. → the thumbnail bridges two worlds the real mrbeast next to his animated version. The old audience recognizes the face, the new format gets a reason to click. → it's a show(mrbeast lab), not a video an animated series where mrbeast is the hero. A prologue, a trailer, episodes. Every part feeds the next one. the part I find most interesting is that this isn't an easy format that everyone can do. you can see it in the gaps: the first video came out 9 months ago, and the next ones only in the last 11 days. 4 long videos. 10m subscribers. a lot of work behind each one. the videos on screen sit between 2.1m and 20m views. would you rather post 4 videos like this or 40 normal ones?
3
1
14
859
Depliny retweeted
I couldn't even imagine Jake Paul video in a talking head format, but today it's a reality. in fact, this could really boost his personal brand and build trust with his audience, since he can reveal more of himself as a person to his audience in this format. is the end goal of these videos to sell a product, or is it something on a larger scale?
4
1
17
894
Depliny retweeted
apple's iphone duo launch hit 3,800,000 people watching live at the same time. here's strategy of every apple's launch on youtube: phase 1: the livestream • 3.8m people watching live at the peak. • 50m views a month later. phase 2: apple's own product videos • iphone duo: 32m views. • iphone 18 pro: 24m views. phase 3: the creators • mkbhd, mrwhosetheboss, dave2d and ijustine all made their own videos about the new iphone. • apple's channel announces the product. The creators tell their audiences what they think. the livestream is the moment. the product videos explain it. the creators convince people. apple understands better than anyone else the power of YouTube for promoting its products.
1
9
538
Depliny retweeted
friendly reminder: start using what has already proven to work
each of these thumbnail concepts outperformed you know what to do with that
1
7
690