Subscribe
Share
Share
Embed
$32,700 in savings. 10 months of runway. non-technical founder building a real SaaS to $10K/month using only AI tools - or going broke trying.
this playlist is my full journey from $1,940 MRR to $10K MRR, filmed in real time with real numbers. the wins. the fuck-ups and the dollars. no code experience. no safety net. just AI tools and a deadline.
new videos every week. subscribe to see how it ends.
# 2026-04-21-threads-growth-is-a-lie
[00:00:00] I got Claude to read 1,996 mega viral posts so you don't have to. Turns out most Threads advice is complete bullshit.
0% of the top 10% of the viral Threads posts used a how to hook
zero.
I posted about this on Threads 10 days ago. 67,800 people saw it. 300 people saved it.
I have spent 1017 days on Threads, posted over 4,500 times, made over $100K from Threads, spent about 3,814 hours there. I went from 17 followers to 27.3 K followers.
So trust me when I say that everything that I thought I knew about Threads growth was a fucking lie.
Lie one is you need an audience to go viral. You do not. The algorithm does not care who you are. It cares about the quality of engagement in the first 60 minutes.
Lie two. Copying what works, works. It does not.
Here's an example of what to do. I have like two followers, so no one will see this, [00:01:00] but I had to tell it. This got 108,000 likes on Threads.
An example of what not to do, which I have literally used this hook several times, how to get your first 1000 followers on Threads in 30 days. The bad version promises instruction. The good version creates immediate tension.
Notice what you should copy here. It's not the exact sentence of the 108,000 like post. It's not the niche. What you're doing is copying the mechanism, so the good hook, it creates a private room feeling. It makes the reader feel like they're about to overhear something they were not supposed to.
Here's what you should do instead, stop opening with instructions, open with an observation, a confession, or a sentence that creates tension before it actually teaches.
The tension is that lots of people end up seeing it and people feel bad, so they feel compelled to go and like it and comment beneath it.
Which leads to finding number two,
Finding two. Mid-length posts die.
This is the simplest rule in the whole video, and somehow it's the hardest one [00:02:00] to follow. If you can say the thing in one sentence, say it in one sentence.
This is what the data says. Posts under 50 words get 65% more likes than posts with between 300 to 600 words.
And it's that middle range between 300 to 600 words. That is the dead zone.
A good example of this, stop wasting your life on other people's drama. That is from Mel Robbins. It's under 30 words and it got 63,000 likes.
Bad example. I have been thinking a lot lately about the complicated emotional and practical reasons people get too invested in drama. And here are five reflections that might help you reconsider your energy.
This bad version makes the reader walk through a hallway. The good version opens the door and just points at the thing.
And this is why short posts feel more confident. Long posts often sound like you are trying to convince yourself, but short posts sound like you already know.
Point being on Threads, certainty travels faster than explanation.
Here's the key takeaway. Write the long version privately, post the sentence that survives after you cut it [00:03:00] twice.
And that brings us to finding three.
Finding three says, two posts max.
Two part Threads outperformed three part Threads by 71%.
One sharp opener, one clean follow up and done.
The second post can add a payoff, but the third post, it usually adds drag.
Interestingly, a four post thread averages, 2,455 likes versus a three post thread, which averages 1,870. So one post is good, two posts is good, four posts is good, but not three.
A good example from the Threadify vault, unpopular opinion. Sometimes the healed version of you is meaner.
Then there was a short follow up post beneath this one, and it got 58,082 likes.
Bad example. Imagine a three post thread. Part one is, here's the belief. Part two is here's the backstory, and part three is why that matters.
The point here is the bad version turns one feeling into what feels like a loading screen, but the good version it hits once, clarifies once, [00:04:00] then gets out of the way.
The second post, it should not start a new idea. It should sharpen the first idea.
Think about it like this. If your second post needs a third to make sense, the thread, it's like a leaky bucket.
So here's what you should do instead. If one post is not enough, use two. If two is not enough, it means your opener is not sharp enough yet, or turn it into a fourth post.
Whatever you do, just don't make it three.
Finding four. Trigger words scream fake.
Bottom performers use two to eight times more emotional triggers than top performers.
Speed words were seven and a half times more common in the bottom 25%. Fear was 3.8 times more. Common pain was 2.1 times more common.
This basically is saying that the post trying the hardest to make people feel something performed the worst.
An example of what to do. Having parents who care for you in adulthood is a gift. 47,600 likes. There's no fake urgency. There's no pain bait. It's just a quiet truth.
A bad example, if you do not fix [00:05:00] this painful mistake today, your growth will die fast. I've been guilty of doing stuff like this so many times.
It doesn't perform well because it performs emotion, whereas the good version contains emotion. That difference is everything
Performed emotion tells the reader how to feel, but contained emotion gives the room to feel it themselves. And the more you push the feeling into the sentence, the more the reader feels the push.
So here's what you should do instead, delete the trigger words and say the truth plainly. If the sentence only works with urgency sprinkled on top, the sentence is not strong enough.
By the way, every good example in this video came from the Threadify Vault. I'll show you the vault properly near the end, and then I'm gonna give you the GitHub repo for the forensic skill so you can get insights like this and make your content creation so much better. Anyway, we've got three more findings.
Let's dive in.
Finding five. CTA has cost you
Adding a call to action cut likes by 12% across 1,944 posts. Things like Save this, follow for [00:06:00] more, comment below every ask is adding friction.
This one hurts because I sell a product and I have written so many needy CTAs. But the data says that those asks often turn, reach into resistance.
An example of what to do. There was this story about a walnut allergy on Threads. It got 123,000 likes. It did not have a follow me or a save this, and there was no check my bio. It was just a story worth sharing.
A bad example if this helped, follow me, save this comment, growth, and send it to a friend. There are so many asks in the Threads posts that it's creating more resistance, and obviously it's asking for four jobs. The good version. On the other hand, it gives the reader one feeling,
But what do you actually do with this if you have some sort of business intention on social media, because you need to make an ask to get people to know about your stuff and to use it. So here is the uncomfortable business lesson.
The viral post's job is reach.
The conversion post's job is conversion. When you [00:07:00] force one post to do both jobs, you often make it worse at both.
Now, the 1 0 1 of all of what we've just said is this. Let viral posts earn attention. Put your stronger CTA somewhere else after trust has already been created. The 2 0 1 is let a viral post earn attention in the first hour or so, and then drop a CTA beneath it. Once the algorithm has picked it up and starts distributing it to more people on the platform, that way you get the best of both worlds.
That is exactly what I did with this viral post that this YouTube video is based on, and at the end you can see the last post doesn't have any CTA to follow me. It just has a T-L-D-R. Then an hour later I dropped a CTA to my free Skool community Full Circle and it got about 30 or 40 people to join and as this guy commented, okay fine, the funnel got me.
Two more findings left.
Finding six questions. Bleed likes,
Posts with questions got 19% fewer likes than posts without them.
Put differently [00:08:00] statement outperformed questions.
This one again really surprised me because the old advice says to ask open questions because questions create comments, and comment is good for the algorithm, right? But the data says most questions just create homework. They create resistance for the reader.
We return to this post from finding one. I have liked two followers, so no one will see this, but I had to tell it statement, not a question. 108,000 likes
Bad example. What is your biggest struggle with content creation right now? This bad version? It hands the reader a worksheet. It gets them to think, okay, well is it that I overthink my posts? Is it the perfectionism? Is it that I feel like an imposter? I don't know. I keep scrolling. Whereas the good version, it hands them a feeling of, oh, I don't want you to have like two followers.
I want to support you. I want to get around you. I'll like it. And I'll say Followed. 'cause that was easy and it was brainless and it didn't require me to think.
Here's the mental model you can use to think about this. If the reader needs to stop and [00:09:00] compose an answer, it's probably a weak question, but if the reader instantly thinks that is me, it's not really a question anymore. It's recognition and that recognition will then get them to go and comment.
So here's what you should do. Instead of asking a question, turn your question into a flag and plant it. So don't ask, do you agree? Say the thing people will want to agree or disagree with.
One finding left. And this is the one I've saved until the end because it is what changes the game for creators who create long form content.
Finding seven. Long form needs blood.
This one has broke my brain about what I thought I knew about Threads because I have written one long form post every day for hundreds of days, and the data says the only long form posts that broke through were raw, personal stories. There was not one educational thread that made it into the top tier of the data.
Now you're probably thinking that there's a bit of a contradiction because you are literally watching an educational YouTube video right now.
The way to sort of reconcile that contradiction is [00:10:00] simple. The educational stuff has its place, but if you want to go mega viral like Mr. Beast, mega viral, educational is not the way to go. It's entertainment. It's content that makes people feel something.
An example of what to do was, my morning routine when I was homeless. It was a four part thread. It got 59,100 likes.
Another one. Started with this hook. I told my therapist one day. It was a nine part Threads and it got 58,900 likes.
A bad example. 12 lessons I learned from studying successful people. Part one of 12.
Notice the difference between the good examples and this bad one. The bad one promises information whereas the good versions, they risk vulnerability.
Sharing what you told your therapist, that's vulnerable. Sharing your morning routine when you are homeless, that is vulnerable.
And that right there is the price of long form on Threads. If you take more time from the reader, you have to give them more humanity back. Information alone is not enough. A long post. It [00:11:00] has to make the reader feel like they got closer to a real person.
Here is the key takeaway. If you go long, go personal, it should feel uncomfortable to share. That's when you know you're on the right track. And if you can't make it personal, keep it under 50 words.
Here is the TLDR.
No how to hooks, observations only.
Keep it under 50 words unless the story is genuinely personal.
If you're writing more than one post in a thread, make it have two posts or four posts, or five, but three is the dead zone.
Don't use fake urgency. There were two to eight times more emotional triggers in the bottom performing posts than the top.
If the goal is virality, don't put a CTA in there. Make viral posts, earn their attention and make conversion posts convert.
If you put questions in your content, you can expect a 19% reduction in likes.
Just say the thing, don't ask it.
And finally, if you go long bleed, be vulnerable. The educational tip Threads are five times overrepresented in [00:12:00] the bottom portion of the data.
The reason all of that works is simple. Threads rewards fast, high quality engagement. This is what I call the golden hour.
Replies matter more than likes, but only when they happen naturally.
And this is good news for you. It means that you don't need a giant audience. You just need a post that creates the right first 60 minutes.
If you can do that, get people to comment on your post, this is signaling to the algorithm that this is going to keep people on the platform for a longer time because it takes time to reply to something and then the algorithm will then go and distribute it to more people on the platform, because more people on the platform means more advertising dollars and more advertising dollars means more money in Zucks pocket.
This is the vault. Every good example we just went over came from here. There are 1,423 in the global vault, which you have access to, and then there are 552 in my own private vault and together that represented the 1,996 posts.
We add to this global vault with posts that [00:13:00] get over 1000 likes or more every single day.
I can search for posts with keywords, so if your niche is triathlons, you can search for triathlon. If your niche is cooking for kids under 10, you can search for cooking.
You can filter them based on long form or short form.
This is how I stop guessing. Every morning I go to the Threadify vault and look for inspiration and go from there. I can even do the two click repurpose, where I click on the little sparkle button, make it focus on Threads growth, which is my thing, i'll put my offer in, which is Full Circle, and then generate. Then it's going to look at the brain, the knowledge of information that it has on me, and then it will use that, plus it's understanding of viral posts to create like a, a baby. Then you go and post it and you might go viral.
And this is what it generated. Honestly, it's a really good post.
So the vault, it shows you what works, and the brain helps you turn those patterns into posts that sound like you.
And that difference really matters. The whole point of this video is not copy viral [00:14:00] posts. Copying is part of the lie. The point is to understand the pattern and then write it in your own voice or just get Threadify to.
So if you wanna stop guessing on Threads, start here. Threadify is free to start. You get the vault and brain for free. Go to www.threadify.app/plans
And one more thing because you made it to the end the gift that I promised you at the start,
This is the AI Content Forensics skill, and the GitHub repo is public.
Initially created this so you could reverse engineer any YouTuber's content strategy with data, not guesswork. And you can use this with Claude Code or Claude Cowork.
And I just updated the skill so it can analyze Threads accounts too.
So you point it at a Threads creator and build a corpus. Pull the patterns, write the thread, and make the visuals all in one prompt.
I am not putting this in the description because this is a thank you for the people who actually watched this whole thing. So pause the screen, take the repo, take the link, use it, remix it, break it, improve it.
And let me know in the comments below who you want me to do [00:15:00] AI Content Forensics on next. I was thinking Chris Williamson or Tim Ferriss, but I wanna get your take. So just drop a comment, put the name and let me know.
And until then, don't be a drop kick, subscribe. Peace.