How To Grow Fast On X

Nikita Beer, who's the head of product at X. So he actually tweeted something and he deleted it, but this is how to grow fast on X. And this actually ties in with, this ties in with content in general. So let me read this out to you. So Nikita says, CT, crypto Twitter, is dying from suicide, not from the algorithm. CT, again, crypto Twitter, ends up wasting all their reach on replying good morning. So GM, right? So crypto Twitter would be GM, we all gonna make it, all these things, right? Hundreds of times. And when they finally post real content like a project announcement, it only gets shown to 3 people.

Let's go back. What's CT?

Crypto Twitter. Okay, Crypto Twitter.

I don't know what that is.

3, 4 years ago, everyone was like, um, they're all— all these acronyms. So if you say GM, like good morning, everyone responds GM, GM, GM, GM, GM, right? Um, or it's like, um, you know, crypto's going down, it's okay guys, we all going to make it, right? So everyone's using these acronyms. But his point is crypto Twitter kills itself because they are wasting their reach. So the point is The average Twitter user only views 20 to 30 posts per day, okay? So crypto Twitter ends up wasting all their reach on replying good morning hundreds of times, and when they finally want to post real content like a project announcement, it only gets shown to 3 people. So the point is, don't try to post in volume, try to post something with a point of view, something that's high quality, and then do it only a couple times. Don't just try to spam Twitter, 'cause people are like, oh, we can spam Twitter. You actually can't. That's his point.

Yeah.

Same thing with LinkedIn.

That makes sense. I figured it's just like most social networks. If you go on Instagram and you post 20 times a day, your reach gets really compressed. And then in the future, when you want to post, the algorithms also are like, hey, your followers don't like engaging with you. We shouldn't show your new stuff, even if it's good, that often. So you got to start rebuilding your trust with these platforms and your followers.

You know, at least with this podcast, our best content is when We have a story and we're just being ourselves. And so I realized yesterday, I was telling you this pre-show, but I was— you're speaking at the summit today. I was speaking at the summit yesterday and I was like, you know what? I'm going to go in with no deck, no preparation, nothing. Okay. I called it something else, but anyway, that's not the point. So the point is I'm going to go in and I just came in with a ton of energy and then I did some screen shares going back and forth and everyone in the chat's like, yeah, yeah, yeah, yeah, yeah. Everyone's going crazy. Right. And then Because Granola was recording that, I took that transcript. I also took a transcript of us just riffing on this, right? Like 120 minutes or so, or yeah, 120 minutes or 80 minutes, 80 minutes. And then I compared it to me talking at a conference like the HubSpot conference. And I was like, okay, which one? You are the best speaker in the world, evaluate me. It's like by far when I'm just no slides, no preparation, it's way better.

Right?

And it's like, but the problem is the content that you present is technical. So you still need to have some slides, but do it differently because when it watched me speak at the HubSpot one, this is, I'm using Gemini here, by the way, because Gemini can see the YouTube, right? It's like, yeah, you're getting it. I can tell that you're getting held back by the slides. And so I'm not saying this style is for everyone. I'm just saying you have to understand what style works for you. And then once your style comes out, that's going to become higher quality content, at least for videos.

Yeah, I agree with that. I think people should just be themselves and not try to prepare too much. When I go and speak, a lot of people have asked me over the years, hey, you've done a decent job speaking. We like hearing you. What's your secret? And I'm like, I'm not really speaking at the audience. I'm just having a conversation. Like, what do you mean? I'm like, I just go up and I just talk. I don't prepare. I'll look at my slides when I'm on stage. Sometimes I do look at them like 5, 10 minutes before I go on stage. My team makes them for me. tell them what to put in the slides or how I want the outline. So, you know, that makes it easier to not have to look at them. And I just have a conversation with people and just talk. And when you're just being yourself and you're not putting pressure, it tends to work out pretty well.

You know what's funny, Neil? I was reading one of the comments with us talking, and then one of the comments was like, Neil's like chamomile tea. And then Eric's like, he took 3 shots of espresso before he started.

Dude, like 4 or 5 people sent that to me.

Oh, you saw that too?

Yeah.

Yeah. I thought, I don't think I sent it to you, but I've realized something because I'm so excited about these things because even at the HubSpot talk, I was going really fast, but I'm like, no, this is me when I'm excited about it. And some people be like, usually for most channels I have to speed it up, but when I land on your channel, Eric, I feel like I have to slow it down and I love it, right? And then one guy posted yesterday, he's like, I like this guy and I don't even need to speed up the video. So I'm not saying that's for everybody. Like Neil has a very calm, collected, it's slower pacing, which is good. But I realized over the years, I'm like, wait, This is just me and it's not going to be for everyone. But you look, when you listen to Marc Andreessen, I heard Marc Andreessen talk today from Andreessen Horowitz. I'm like, oh my God, it's on 1.5x and he's like on 3x and I'm like, but that's him. He's not going to change. And I'm like, that works for me.

Wait, it's on 1.5x?

I had it on 1.5x and it sounded like 3x because he talks so fast and I can't understand him.

Oh, so you went back to normal.

I had to go back to normal. So my point is, you have, again, you're not going to develop your style immediately, especially if you're starting out right now. It might take you a couple of years, but once you find something that works for you and it feels good, Double down on it.

Yes. And when Eric talks, sometimes I'll tell him to slow it down because it's, if he talks too fast, I think people will miss the point he's trying to make. And it's valuable and I want people to hear, but majority of the time I never tell him to slow down. Why? Because when he's talking fast, it usually means he's passionate and he loves what he's talking about right then and there. It comes off differently than if I told you to slow down. And that's why I only tell you to slow down if I'm starting to miss things, because that means people may be missing things.

And to be fair, when Neil does that, it's It's to help the audience, right? More than anything, it's to help the audience. That's helpful. So that's why I don't get offended by it.

So correct. And when Eric wants to talk fast and he's passionate and people are going to get it no matter what, I let him go because the passion he has, it just comes off better in my opinion when you're just yourself. And I don't think, you know, like back in the day, people are like, oh yeah, you know, Zuck, he's media trained. Oh, look at him now. If you look at Zuck today, I don't think this is the media trained Zuck. Would you agree with this? This is Zuck wearing a chain and just being himself.

He wants to do Brazilian jiu-jitsu, go do it. You're going to break your— I think he tore his ACL. Cool. But he's being him.

Yes. And people relate more in today's world, especially with that, than these robotic trained people. And the reason I say that is in this AI world, it's really screwing people up. For example, I'm posting something tomorrow. This episode doesn't get posted tomorrow. Check this out. Okay. All right. So I got this chart. Do people know when they're viewing AI-generated social media content? We did this with 100 people. We showed them AI content and non-AI content.

Okay.

So the first batch, only human-generated content. How many times or what percentage of times do you think people thought it was AI-generated?

Zero?

24%.

Wow.

So they thought when we showed them human-generated content, this was only human-generated content, 24% of the time they thought it was AI-generated.

That's interesting.

Okay, then we showed them AI-generated content. We didn't tell them it was AI, right? We're just showing them content and we're classifying it on it. When we showed them AI-generated content, what percentage of the time do you think they thought it was AI?

24%.

36%.

Okay.

So majority of the time, 64% for AI-generated, 76% for non-AI-generated, majority of the time humans believe content is human-created, whether there's AI in there. And if you actually think about it, A percentage of the time, roughly 25% of the time, they think content is AI-generated.

Dude, we're not going to be able to tell anymore.

Yeah, it's getting harder and harder.

Did you see that? Have you seen that monk? There's a monk in a temple and he's always spewing wisdom and then it eventually goes to like a course. He's like, you know, it's not about money, blah, blah, blah. It's like an AI-generated monk, right? And then there's like a link to go buy his course.

I did not see that, but I saw one video of a guy sitting on a chair and he's just talking and moving his head.

And it constantly changes from a woman to a kid to adult to a guru.

Like, he changes to all these different characters and spewing different stuff just to show you the power of AI. It's crazy.

There's some on Twitter now, you can see there's like guys that are moving around, but the avatar is a woman and you actually can't tell anymore. And so I think what's gonna happen is you're gonna see a lot of virtual, you know, OnlyFans and it's just, you know, guys behind the scenes playing it, so. Yeah, that's what we're going to see. If you're in B2B, here's the deal. LinkedIn account-based marketing sucks, but it doesn't have to. With Karrot, it allows you to create personalized ads towards your customers and also personalized landing pages that align with the ads. So you're going to have higher conversions and influence pipeline reporting to prove that your ABM works. So if you want to learn more, just go to karrot.ai and you can join the waitlist. Again, it's karrot.ai. And we'll see you on the other side. Yeah. By the way, okay, but I do think you just kind of mentioned this and you have this highlighted. So I was talking to my team about this. So we mentioned, okay, brand's a differentiator. Live events are like in-person is a differentiator. Having taste is a differentiator. But it's like, okay, well, if people can code, we can build all these things. What is the ultimate differentiator when it comes to building things? It's design. And I'm like, holy crap, we need to get really good designers because they have tastes and they already— they can build whatever they want now too. But you and I both know, like, I wouldn't say, you know, I have the best design taste. I don't think we've had the— historically we've had the best designers for your company or mine.

No.

But I do think we're going to be looking for some of the best designers in the next year or two.

That's what I think. I have some designers that I'm really happy with on the product end. I always really struggle with website design because what my executives want versus what I want are two different things. And then I let them go and do amok with the website and then they look at conversions and we're optimizing. Then I go and change stuff and then Saulo goes and changes stuff. And what we found is for enterprise-based leads and business, like trying to close like 7, 8-figure accounts, a lot of the design does not matter one bit.

Yep. Yep. I mean, oftentimes they just want to get to the point, so.

Yeah. And what I've learned through this, I'm like, well, If revenue is growing and my team can constantly hit numbers, do I really want to have the fight with my management team on what design looks better versus worse when we A/B test this and something may look more pretty, but it doesn't really impact revenue? It's just like, all right, I'm not going to have this battle. Go do whatever you want.

But I would say, so I agree with that piece. I would also say that the best designers are typically very, very opinionated and they push back a lot. And we did this with the— so we just launched the ClickFlow site today, right? If you look at the landing page, And we were talking, oh, we're like, oh, we don't have a new logo. I was like, okay, that's what my, my, my— I was like, hey, just put the original logo we had on the site and see, and then it's going to piss off design. It pissed off the designer. He redesigned it and like, problem solved.

The new design's good.

Yeah, it's pretty good. And it's better than a single green website.

So. Yeah. The only problem you're going to have is your CTAs. A lot of them blend in because of the coloring.

Yeah. Don't worry. We'll change it.

Yeah. But other than that, the design is great.

But see, that logo used to be, it said ClickFunnels orange before. And it pissed him off so much. And then that, that the, the, the, the lines, the funnel lines on it, like that wasn't there before.

My problem with design is I tend to optimize for conversions more than anything else.

Yeah. So like, so you'll be at odds with designers all the time.

I'm at odds with designers and I tend to control what gets shown for, let's call it half the homepage or majority of the homepage. And then I just don't care. So like I'll optimize for the areas that I know that are driving conversions and then they can do whatever they want with the rest.

That's what happens when you're very focused. So we'll leave it at that. So you had the death of rollups highlighted, or did you want to go with the distribution channels? Actually, no, hold on.

I didn't highlight any of them. My mouse is just moving.

Well, let's— I want to continue. So we were talking about design, but Shopify, you saw Shopify, Google, Walmart, all these people have partnered on basically agentic commerce with UCP. You want to talk about that?

Let's go for it. What's this tweet?

Okay. So—

There's a tweet by— oh, this is Tobi from Shopify.

Go for it.

Uh, I'll let you read it.

Go for it. So Tobi, CEO of Shopify, by the way, hey, before we go into the Shopify one, Tobi tweeted this recently. He said that in the past 3 weeks, this is a week ago, in the past 3 weeks, he's shipped more code than he has in the past 10 years using Cloud Code. Yeah. So that's high leverage for him.

Dude, I haven't heard too many people talk about Cursor. Everyone's been talking about Cloud Code.

I use Cursor within Cloud Code. Sorry, I use Cloud Code within Cursor all day. And that's what Yaniv does as well. So yeah, I do, and you have a lot more options.

So you're just using both?

I'm using, on one, on the far end, I use it as a data analyst with all my MCPs. I'm just asking, querying my data questions. On one end, I'm using it to build stuff, and I have multiple tabs to build things. Think of it as a browser.

Can't Cloud Code do a lot of the stuff that Cursor does?

But in Cloud Code, you know MS-DOS, right? Like a terminal? Yeah, you're using terminal. You have one window open, but I have that, and then I have the output, the .md files, which I can basically edit if I want. the email copy and then save it in there. It's not that easy to do inside of terminal. So it's actually better to have it live within like an ID, like a cursor.

So they'll eventually fix that and then you won't need cursor.

Yes, exactly.

It's just a question of when.

Yes. And they did talk about that on the 20VC. It was Harry Stebbings, Jason Lemkin, and another guy. They're talking about, it's like, what happens? And I think their whole stance on it is like, everyone's going to have, they're going to be able to survive. We'll see what happens. I'm just glad I'm not in the business. It's tough. Okay. So back to this UCP thing. So guys, MCP is Model Context Protocol. It's like APIs where you can talk to your products. You can have the LLMs talk to your products, like have ChatGPT talk to your HubSpot, for example, and ask it questions. UCP is Universal Commerce Protocol, which they developed, they co-developed with Google. So Google and Shopify partnership, it will make— UCP will make it faster for agents and retailers to integrate. It's open by default, so platforms and agents can use UCP to start transacting with any merchant. Any merchant, yeah. Major retailers are already using it. Finally, agents can handle everything from discovery to fulfillment and support things like discounts, subscriptions, and loyalty programs. We've accounted for all types of commerce. So imagine this, if you're using AI mode and you're searching for best luggage, you're going to be able to buy it directly from within AI mode. That's one example.

I think this is going to be huge. We're going to see this as, you know, more and more adoption comes, and I think it's going to end up happening really soon, and it's going to affect the way that people are shopping, but not fully affect it because I think people are also going to shop the way that they're used to right now. I think it's going to be a combination of both. What I'm really waiting for when it comes to shopping is you're able to put in a specific requirement into Amazon because we are trained to just go buy on Amazon. You agree with this, right? For like most products, you still buy on Amazon.

Just one click and one click add to order.

Yes. The problem with Amazon And let's go with the example of a jacket because we're both wearing sweaters or jackets, whatever you want to call it. Let's say we're going to, um, or this is a perfect example. I got my son ready for school this morning. The weather during this morning was in the 50 degrees. It gets into the high 70s in the afternoon. It'll be 80 today. Yeah. So it becomes a little tricky on how you want to dress the child. You don't want them to be cold in the morning. You don't want to be hot later on. And this happens a lot in Los Angeles.

Mm-hmm.

I wish I could just go on Amazon and type in, looking for kid pants or need kids' pants, needs to keep you warm when it's cold, but the same pant needs to keep you cool when it's hot, needs to adapt in temperature. What's the best pants for kids? Plays in dirt, gets dirty, goes on slides. And then it just says, Here are the best pants for this scenario. I'm waiting for that to happen. I think that's more useful than all this agentic commerce stuff that people are showcasing right now with the technology, because we buy right now by doing research and typing instead of just saying, hey, go scour not just the web, but all of Amazon, all the reviews, and tell me what is the best product for my needs.

I wonder, as you say this, I actually want to pull up a graph to see how fast, how quickly it took for people to adopt to mobile. Like smartphones, right? When the iPhone came out, let's just say 2010, and by what is it, like 2015, 2017, maybe the whole world changed. Do you think it's maybe 7 years of mass adoption? Does that sound about right?

Sure, that sounds about right. But for me, it's not just even like these devices. If you go on ChatGPT or you're in AI mode, And it tells you the product to buy and shows you, would you buy right then and there? Would you go to Amazon? Honestly?

No, right now I would still have our default behavior, but I'm just wondering, the reason I asked that, posed that question is how long does it take for human behavior to change en masse? And you can see iPhone actually launched in 2007. It exploded into mainstream 2012 to 2013. So let's call it 6 years or so, over 80% penetration by 2016. So let's just call it 9 years or so. So my point is, Is validating what you're saying. It's human behavior takes a while to change.

It does take time to change. Now, would I be open to purchasing in AI mode or would I be willing to purchase through ChatGPT? I would, but I want the luxury that Amazon has to offer. And with a lot of Shopify stores there, you don't get that luxury. I order from Amazon, dude, sometimes the same day I'm getting the product, right? I order from Shopify, I'll get a product in a week.

I think Amazon's going to copy this. That's it.

I agree. And then I'll be down to do it within Amazon. That's what I'm getting at. Because the concept of AI mode, you're asking the specific questions that I would be asking, like on my son's pants. But I want to do that in Amazon because we're ingrained to go to Amazon because of their fulfillment network. Their logistics is better than anyone else and you get products right away. I don't want to wait a week or two. So I want Amazon to implement this. That will change shopping drastically more than I think most people think.

I bet you they're doing it already because all this Claude code stuff, because every, like, by the way, the Claude, this Claude for Cowork thing, there's like, oh, it's built 100% on Claude code. And then the Claude, the creator of Claude code is like, yeah, all my code's written by Claude code. Right. So there's no way they're not doing this because if our engineers are, you know, 5, 10, 15 times faster, they already have a bunch of people working on it, guaranteed.

Yes.

Alright guys, we'll catch you later. Goodbye.