JACK NEEL · EXTRACTED

Short-Form Expert: "Going Viral Isn't Luck." I Make $1M/Month With 31 Second Videos

The loop trick, the competitor copy method, and why scripting is 80% of the work. Most creators obsess over consistency and quantity. The actual lever is quality, structure, and knowing exactly where your video loses people.

▶393K views on YouTube

"If you understand the fundamentals of how to go viral on social media, you can turn anything viral. I did it on Snapchat. I did it on YouTube shorts. I did it on software. I did it on a personal brand. I've done it on Minecraft channels. People over complicate this way too much." — Daniel Baton
Daniel Baton, on the episode

Daniel Baton started running faceless content channels at 12, was making half a million dollars a month on Snapchat at 15, and co-founded Crayo, a short-form video editing software, before graduating high school. The pop framing of short-form success is that it requires constant output and luck. The actual operating system Baton runs is more surgical: find a competitor already winning, identify every dimension where you can beat them by even a small margin, and compound that edge video by video. This protocol pulls three of the highest-leverage plays from the conversation, each with a specific mechanism behind it.

Tactic 01

Psycho-Analyze The Drop-Off, Then Remove The Section

Most creators treat retention graphs as a report card. Baton treats them as a diagnostic tool with a specific repair loop attached. After every upload, he goes into YouTube Analytics, finds the exact second where viewers drop off, returns to the script, and asks a precise set of questions: Did I drag it on too long? Did I give too much context? Did I not go deep enough? Did I fail to visually represent what I was saying? He does not guess at the answer. He forms a hypothesis, changes one thing, and tests it on the next video. The compounding effect is the point. Each video carries forward the fix from the last one. Over 30 days of daily uploads, you are not posting the same video 30 times. You are running 30 iterations of an improving formula. Baton calls this psycho-analyzing the guts of the video, and it is what separates creators who plateau at a thousand views from ones who break through to a million. The loop is cheap: check stats the day after posting, identify the single biggest drop-off, go back to the script, form one hypothesis, apply the fix.

The play

After your next upload, open YouTube Analytics and find the retention graph. Locate the first significant drop-off point. Go back to your script and identify what was said or shown at that exact second. Form one hypothesis about why viewers left, whether the hook dragged, the context confused, or the visual did not match the audio. Change only that one thing in your next video. Repeat this after every upload for 30 days. Your retention line will flatten toward the end of the video faster than any other method because you are building on a real signal every time.

Tactic 02

The Competitor Copy Method: Three Upgrade Metrics

Baton does not start channels from scratch. He starts them from proof. Before posting a single video, he finds a channel in his target niche that is already averaging over a million views per video, downloads their top 10 most popular videos, transcribes the scripts, and highlights the hook, the supporting sentence, the rehook, and the payoff. Then he identifies three specific upgrade vectors: can he make the videos longer, can he raise the visual quality, and can he improve the concept angle. The length point is backed by a concrete mechanism. YouTube places more ads on longer videos, so it has a direct financial incentive to push them. A channel posting 25-second videos is leaving distribution on the table. Baton goes to 45 seconds on the same concept and captures both the audience that wanted more and the algorithmic push that rewards the extra watch time. The quality upgrades are often small. If the competitor uses white subtitles throughout, Baton makes impact words like hate, scam, or declined turn red. If the competitor's clips are in 720p and still pulling a million views, he shoots in 1080p. These are not dramatic changes. They are deliberate marginal improvements that compound into a distribution edge.

The play

Pick one faceless niche where you can find a channel averaging over 800,000 views per video. Download their five most popular videos. Transcribe the scripts and highlight the hook, the first supporting sentence, the rehook, and the payoff. Then write down three specific upgrades: one length change, adding at least 10 to 15 seconds to their average duration, one visual quality change, and one concept or angle change. Post your first video using their structure with all three upgrades applied. Check the swipe rate after 24 hours and aim for above 78 percent, which Baton cites as the threshold where videos break through to multi-million view counts.

Tactic 03

The Loop Trick: Engineering 200 Percent Retention

YouTube weights the beginning of a video heavily when deciding how widely to distribute it. Baton found a way to manufacture a retention spike at the start that signals to YouTube the video is performing at an extraordinary level. He calls it the loop. At the end of a short, instead of letting the video simply end, he adds a sentence that feeds back into the opening hook. His example: a video about Speed hating Mr. Beast ends with the line, so that's why Speed hates Mr. Beast, so why? That question is also the opening line. Viewers who hit the end of the video are pulled back into the first two seconds before they realize the video has ended. The result shows up in analytics as 150 to 200 percent retention on the opening seconds, meaning more people are watching the beginning than the total initial audience. That number tells YouTube the video is unusually compelling and triggers a wider push. The mechanical execution is simple: take the first clip of your video, extend it slightly, and attach it to the end of the timeline so the closing shot and the opening shot are the same continuous clip. For faceless content, this is nearly invisible to the viewer. The sentence that closes the video should be either a question that is also the opening question, or a phrase that sounds like the start of a new thought rather than an ending.

The play

At the end of your next YouTube short, write a closing line that is also a question matching your opening hook. Then in your editing software, copy your first clip, extend it by one to two seconds, and paste it at the very end of the timeline. The video should appear to continue rather than end. When YouTube Analytics shows 150 to 200 percent retention on the first two seconds after a few days, that spike is the algorithm seeing the loop and reading it as exceptional viewer interest. Baton reports this method alone has contributed to videos breaking out to multi-million view counts that would otherwise have plateaued.

Tactic 04

Find Editors On Discord, Not Job Boards

Most people trying to build a short form channel get stuck at the hiring stage. They post on Upwork, scroll through Fiverr, and end up with editors who produce generic work at generic prices. Daniel's answer is Discord, specifically the editing communities that have been quietly filling this role for years. "The best way to find talent for YouTube shorts is to go on Discord servers," he says. "There are a ton of editing Discord servers. And this could be for any type of content. You could be a personal brand. You could be a faceless creator." A quick search for "editing community Discord servers" pulls up lists of communities with anywhere from a few thousand to hundreds of thousands of members. The talent is there, it just isn't where most people look. What makes this work is the structure of how you bring editors in. Daniel doesn't post a flat rate and wait. He posts a job listing with a ceiling, something like "potential to pay up to $10 per video," which attracts applicants, then negotiates the actual starting rate in the DMs. From there the deal is performance-based: start at $2 to $3 per video, move to $5 once the channel hits a million views per video on average, move to $10 once it consistently clears five million. The editor has a clear path. You have cost control while you're still figuring out what works. Both sides have skin in the outcome. "You can start these people at the price you want them to be," he says, "and essentially I'm going kind of in the weeds of how to create the team here, but this is what it takes." The whole point is that the editor grows with the channel instead of costing you full rate on videos that haven't proven themselves yet. There's a deeper reason Discord works beyond just price. The people in these servers are often talented editors who don't have enough consistent work. They're not charging agency rates because they don't have agency pipelines. Daniel's approach is to find someone with potential, get them on calls while they edit, and shape the output in real time. "You sit on calls with these people for hours a day, they edit videos on the call with you. You change what you don't like whilst you're editing." The Discord hire isn't a freelancer you hand a brief to and wait. It's someone you build alongside, which is how you get channel-specific quality without paying for a polished stranger who doesn't know your format.

The play

Search "editing community Discord servers" in any browser and join two or three of the larger servers that match your content niche. Post a job listing stating the role, the daily upload requirement, and a potential ceiling rate ("up to $10 per video"). When applicants DM you, open at $2 to $3 per video and lay out the performance tiers in writing: $5 per video once the channel averages one million views, $10 per video once it averages five million. Get on a screen-share call for the first few edits and correct in real time rather than sending notes after. This keeps the feedback loop tight and builds the editor into your format instead of theirs.

Tactic 05

Tie The CTA To The Viewer's Identity

Most creators ask people to subscribe as a favor. Daniel treats it as a mirror. The idea is that a subscribe request framed around the viewer's own identity, what they already know, feel, or believe, removes the sense that they're doing something for you. They're doing something for themselves. The example he keeps coming back to is a video about Mr. Beast's secret janitor, which slowly built toward revealing the janitor was Chandler. The whole script withheld the name until the final line. The payoff was: 'If you knew it was him all along, subscribe.' Not 'subscribe for more content.' Not 'hit that bell.' The ask was an extension of the story. As Daniel explains it, 'people have this confidence about them where it's like, yeah, I knew who it was the entire time. Let me hit subscribe.' The viewer isn't helping you. They're confirming something about themselves. The same logic scaled to other videos. 'If you'd get mad at Mr. Beast, subscribe. If you thought Mr. Beast scammed you, subscribe.' Each variation puts the viewer inside the narrative and makes the conversion feel like a natural conclusion rather than an interruption. Daniel ran this on faceless channels, personal brand videos, and short form content across platforms. The through-line was always the same: 'I always relate the conversion to the viewer, because why would I relate it to myself? That's being selfish, and the viewer has no real need to help me.'

The play

On your next video, write the CTA last, after the script is done. Look at the payoff of the video, what the viewer just learned or felt, and write a subscribe or follow line that only makes sense if they were paying attention. Use the format: 'If you [specific reaction tied to the content], subscribe.' The line should be unsettling to someone who didn't watch, and obvious to someone who did. Do not add a second CTA. One identity-linked line at the end of the payoff outperforms any number of generic mid-video asks.

Tactic 06

Use 1M+ Average Views Per Video As Your Niche Entry Signal

Most people pick a niche based on what they're interested in or what seems popular. Daniel Baton uses a harder filter. Before he starts a channel, he looks for niches where existing channels are already averaging over a million views per video. Not total views on the channel. Average views per video. The math is simple: if a channel has 30 uploads and 40 million total views, that clears the bar. "If there are channels that have over a million views on average per video, that is my metric. It can be around 800,000 and above, but that's what I look at." The logic behind the number is that it removes guesswork from the niche selection step entirely. A channel hitting those averages has already proven that the audience exists, that the format works, and that the algorithm is willing to push the content. You are not betting on whether a niche will perform. You are confirming that it already does. From there, the job shifts to finding your edge over whoever is already succeeding in that space, whether that is longer videos, higher resolution, better scripting, or a more controversial angle. But none of that work matters if the niche itself is a dead end, and the 1M average cuts that risk before you spend a single hour on production. Daniel is explicit that failing this test means you should not enter the niche at all. "If you don't know how to edit as good as this person, you don't know how to find an editor as good as this person, you don't have an edge. So don't try." The threshold is not a soft suggestion. It is a go or no-go check. If no channel in a given niche is clearing 800K to 1M per video on average, the niche has not proven itself and your time is better spent finding one that has.

The play

Go to YouTube Shorts and search for channels in the niche you are considering. Pull up three to five channels and divide their total view count by their number of uploads. If the average lands consistently above 800K to 1M views per video, the niche passes. If no channel clears that threshold, move on and test a different niche. Run this check before writing a single script or hiring a single editor. Once a niche passes, download the top five to ten videos from the strongest channel in that space and transcribe their scripts. Those transcriptions become your baseline for what the format, hook structure, and pacing need to look like before you improve on them.

Tactic 07

Track Swipe Rate As Your Primary Metric

Most creators obsess over view count and subscriber growth. Daniel points to a different number first: swipe rate. On YouTube Shorts, swipe rate tells you how many people chose to keep watching versus swiped away, and it is determined almost entirely by the first two to three seconds of your video. "Swipe rate is a super important metric," he says. "It tells you how many people swipe away versus watch your video. A swipe rate is made up of like the first two to three seconds. And basically what it's made up of is your hook and your visual. They don't care about the title. They don't care about the description. They don't care about any of that." The threshold matters here because it gives you something to optimize toward before a video has accumulated enough views to read retention meaningfully. Daniel has seen videos with a 77 to 78 percent swipe rate reach ten million views. Below that band, the algorithm is getting a clear signal that viewers are leaving before the content even starts, which means the problem is not the script or the editing but the opening frame and the spoken hook. You can have a perfect thirty-second video that never gets seen because the first two seconds did not clear the bar. This shifts the diagnostic question. When a video underperforms, the first thing to check is not where retention dips in the middle, it is whether the video even survived the opening. A low swipe rate means the viewer made a decision to leave before they heard the second sentence. That is a hook and visual problem, not a scripting problem, and fixing it is faster than rewriting an entire script.

The play

After each video posts, check YouTube Analytics and find the swipe rate before you look at anything else. Your target is 78 percent or above. If a video comes in below 77 percent, go back and watch only the first two to three seconds with the sound off. Ask whether the opening visual alone is interesting enough to stop someone mid-scroll. Then listen to the first spoken line and ask whether it creates a clear, unresolved question in the viewer's mind. Rewrite or reshoot the opening before you post the next video. Do not adjust the middle of the script until swipe rate is consistently above the threshold, because retention data from a video that lost half its audience in the first two seconds is not reliable signal about what happens after.

Tactic 08

Go Longer On The Same Concept

Most faceless YouTube Shorts channels default to 25 or 30 seconds. Daniel's observation is that this is leaving algorithmic favor on the table, not because shorter videos are bad, but because slightly longer videos on the exact same concept give YouTube a reason to push them harder. The platform places ads on videos. More watch time means more ad inventory, and YouTube responds to that incentive by distributing longer videos more broadly. The fix isn't to pad your content or drag out a script that works at 30 seconds. It's to take a concept a competitor is already winning with and ask whether you can give the audience more of what they came for. As Daniel put it, 'I see a lot of really good channels posting like 25 to 30 second videos, which is fine, but I've gone in and then gone to 45 second videos on the same concept. This gives that audience more time to watch content, which they want to do. They want to absorb more content that they actually like.' The audience is already engaged with the topic. Giving them ten or fifteen more seconds of it isn't a burden, it's a reward. The practical implication is that length becomes one of your three upgrade levers when you study a competitor channel. If they're at 25 seconds, you go to 45. If they're at 40, you go to 60. You aren't reinventing the concept or the niche. You're giving YouTube more surface area to monetize and giving the viewer a slightly fuller version of something they already want to watch.

The play

Find a competitor channel in your niche posting shorts in the 25 to 30 second range and getting consistent views. Take their next strong concept and write your version at 45 seconds, adding depth to the middle section rather than stretching the hook or the payoff. Post it and compare your retention curve and swipe rate against theirs. Longer watch time gives YouTube more room for ad placement, which is the platform's direct incentive to push your video over the shorter version.

Tactic 09

Color-Code Your Impact Words

Most faceless channels use a single subtitle color throughout the entire video. Daniel noticed this pattern while studying competitors and saw an immediate way to stand out. The fix costs nothing and takes seconds to apply. His observation is simple: when every word looks the same, no word feels important. But when a viewer sees the word "hate" or "scam" flash red in the middle of a sea of white text, their brain registers a visual signal that something matters here. "Whenever I say a word like hate or scam or one of those impact words, I would make it red," he says. "I would just add any little inputs I could to a specific video to try to give the viewer more visually." This is what he means when he talks about the gap between copying a competitor and beating them. The competitor has already proven the concept works. Your job is to find the small visual details they missed. Color-coded subtitles are one of those details: easy to overlook, hard to un-notice once you start doing it.

The play

Go through your subtitle file before export. Identify every high-emotion or high-stakes word, words like hate, scam, exposed, fired, banned, lie, and change those words to red while leaving the rest white. Do this on your next video and check whether your swipe rate improves versus the previous video. The mechanic is simple: the eye is drawn to color contrast, so viewers who might have swiped stay a beat longer, and that beat is often enough to hook them into the script.

Tactic 10

Hook, Support, Context, Rehook, Payoff

Daniel is direct about where most creators fail: they find a concept that could go viral, then script it badly and wonder why nothing happens. The scripting structure he uses on every video, whether it runs 20 seconds or 10 minutes, follows a specific sequence. The hook gets the viewer to understand what the concept is about without revealing the payoff. Then comes a supporting hook that reinforces the premise. Then context. Then a rehook, which is the moment you pose the question that makes dropping off feel impossible. He gives a clean example. "I show Speed hates Mr. Beast. Last week, I show Speed competed in a Mr. Beast tournament to win $10 million." That is the setup. You add context around what happened in the challenge, and then right before the payoff you ask: but why would this make IShowSpeed hate Mr. Beast? The viewer is now in a state of needing the answer. The payoff only lands because of what came before it. Without the rehook, the context is just filler. With it, the context becomes the tension that makes the payoff matter. This is why Daniel calls scripting "genuinely 80% of the work." A viral concept with a bad script stalls. A decent concept with a well-structured script has a real shot. The mechanics are learnable and repeatable, and they apply whether the topic is celebrity drama, finance commentary, or a 3D science channel. The shape of the story is the same each time: hint at what they came for, build tension around it, make them feel like they almost have the answer, then deliver.

The play

Write your next script in four labeled sections before you record or edit anything. Section one is the hook: state what the video is about without revealing the outcome. Section two is the supporting hook: add a second fact or detail that raises the stakes. Section three is the context: give the backstory or background the viewer needs. Before you write the payoff, write the rehook as a direct question to the viewer, something like "but why would this make him hate Mr. Beast?" Then write the payoff as the final section. If the rehook does not create a genuine moment of tension, rewrite it until it does. Videos built this way hold retention because the viewer is always one step behind the answer.

Tactic 11

Prompt AI Voiceover Sentence By Sentence

Most creators who use AI voiceover generate a full paragraph at once, let it run, and accept whatever tonality the model produces. Daniel does the opposite. He breaks the script into fragments, sometimes three words, sometimes a single sentence, and generates each piece separately. The goal is control: a paragraph gives the AI too much room to make its own decisions about pacing and pitch, and those decisions are rarely what the video needs. The reasoning is straightforward. Short form video lives or dies on how specific words land. Daniel describes it plainly: you need to emphasize certain words to drive a point, to keep people watching, to make the moment feel urgent or funny or serious. When you hand a whole paragraph to the AI, it averages the emotion across twenty words and you get something flat. When you give it three words, the only thing it can do is nail those three words, and if it doesn't, you regenerate just that fragment without touching anything else. He regenerates individual chunks around a hundred times before he settles on what ships. This also solves the coherence problem that comes with AI voice. Because each segment is generated to match the specific emotional cue of that line, the final audio reads as intentional rather than robotic. The listener hears a voice that gets quieter before a reveal, or sharper on a dollar figure, because each of those moments was dialed in individually rather than left to the model's defaults.

The play

Write your next AI-voiced script with a line break between every sentence or every major phrase. Generate the voiceover one segment at a time, not the full script at once. Regenerate any segment that doesn't land tonally before moving to the next. Budget extra time for this: Daniel regenerates around a hundred times across a single video. The result is an AI voice that sounds like it understood the script, which keeps viewers from clocking it as robotic and swiping away.

Tactic 12

Build On YouTube Shorts, Not TikTok

The single thing Daniel points to when comparing platforms is consistency. On TikTok, even if you follow the exact same format you used on every previous video, the platform can demonetize individual videos without explanation. "You could quite literally follow the same format you did on every other video, but they'll just pick and choose certain videos to demonetize and essentially that just kills your TikTok page." The advertised $1 RPM on TikTok's creativity program collapses in practice. "Effectively though, almost everyone I've spoken to has ever said that they're like 10 cents or 20 cents or 30 cents." YouTube Shorts runs differently. If you are in the YouTube Partner Program, every video is monetized from the moment it goes live. You know what you are getting. There is no lottery. The RPM trajectory matters as much as the current number. A year and a half ago, YouTube Shorts was paying 12 to 15 cents per thousand views. It is now consistently in the 30 to 40 cent range. Daniel's projection is that it reaches a dollar within a year to a year and a half, and unlike TikTok's advertised dollar figure, it will be a real dollar applied channel-wide, not 20 cents on one video and 40 on another depending on platform mood. That directional movement is the investment case for building on YouTube Shorts now rather than waiting.

The play

Before you post your next short on TikTok, check whether you are in the YouTube Partner Program. If you are, shift your primary upload target to YouTube Shorts for the next 30 days. Post the same content you would have posted on TikTok. At the end of the month, pull your actual RPM from YouTube Analytics and compare it to your real TikTok creator program payouts from the same period. You are not just chasing the higher number; you are building on a platform where every video you make is monetized at a rising rate, rather than one where the platform decides video by video whether you get paid at all.

Your action plan

the plays, one checklist.

01
Psycho-Analyze The Drop-Off, Then Remove The Section
Check your retention graph the day after every upload. Find the drop-off second, go back to the script, form one hypothesis about why people left, and fix only that in the next video. Do this for 30 days straight.
02
The Competitor Copy Method: Three Upgrade Metrics
Find a channel in your niche averaging over 800k views per video. Transcribe their top 5 scripts. Then beat them on three dimensions: longer runtime, higher visual quality, and a sharper concept angle. Post using their structure with all three upgrades applied.
03
The Loop Trick: Engineering 200 Percent Retention
At the end of your next short, add a line that loops back to your opening hook, a question or phrase that makes the ending sound like the beginning. Then attach your first clip to the end of the timeline so the video appears to continue seamlessly. This creates a 150 to 200 percent retention spike in the opening seconds, which YouTube reads as an extraordinary signal and rewards with wider distribution.
04
Find Editors On Discord, Not Job Boards
Search "editing community Discord servers," join two or three large ones, and post a job listing with a performance-based rate structure: $2 to $3 per video to start, $5 at one million average views, $10 at five million. Edit on screen-share calls together at first so the feedback loop stays tight and the editor learns your format from day one.
05
Tie The CTA To The Viewer's Identity
Write your CTA after the script is finished. Make it a statement only a viewer who watched would understand: 'If you knew it was him, subscribe.' One identity-linked line at the payoff beats any generic mid-video ask.
06
Use 1M+ Average Views Per Video As Your Niche Entry Signal
Divide a channel's total views by its upload count. If the average sits above 800K to 1M, the niche is proven. If nothing in the space clears that number, pick a different niche. Run this check before you write one word of a script.
07
Track Swipe Rate As Your Primary Metric
Check swipe rate before any other stat. Target 78 percent or above. If a video misses that threshold, go back and watch only the first two to three seconds with sound off. Fix the opening visual and hook before touching anything else in the script.
08
Go Longer On The Same Concept
Take a competitor's winning concept and build a 45-second version instead of their 25-30 second one. Add depth to the middle, not the hook or payoff. More watch time gives YouTube more ad inventory, and that is the direct incentive for the algorithm to push your video over theirs.
09
Color-Code Your Impact Words
Before exporting your next video, find every high-emotion word in your subtitles and turn it red while the rest stays white. Check your swipe rate against the previous video.
10
Hook, Support, Context, Rehook, Payoff
Write your script in five labeled sections: hook (hint, no reveal), supporting hook (raise the stakes), context (backstory), rehook (direct question to the viewer that creates tension), then payoff. The rehook is the hinge, if it does not make dropping off feel costly, rewrite it before you shoot anything.
11
Prompt AI Voiceover Sentence By Sentence
Break your script into sentence-level or phrase-level fragments before generating AI voiceover. Produce each segment separately and regenerate until the tone matches the emotional intent of that line. This gives you precise control over pitch and pacing instead of letting the model average the emotion across a full paragraph.
12
Build On YouTube Shorts, Not TikTok
Shift your next 30 days of short-form uploads to YouTube Shorts. Check your YouTube Partner Program status first. At month end, pull your actual RPM from YouTube Analytics and compare it to your real TikTok payouts. You are building on a platform where every video is monetized at a climbing rate, not one that picks and chooses which videos earn.