- Duration: 71 min
- Platform: meet
- Source: Read.ai 01JVQGS96EK9CWEW2VPK6CDYY0 (https://app.read.ai/analytics/meetings/01JVQGS96EK9CWEW2VPK6CDYY0)
Transcript
[Michael Houck]: meeting, internal meeting, that I’ve been on in, I don’t know, like months. Yeah. No, this is good. [Blake Emal]: We’ve got a crowd. We do. Probably a couple more people coming, I’d guess. [Sarah Parr]: Hey, Sarah. [Sarah Parr]: Hey, how’s it going? Good, good. [Michael Houck]: Yeah, let’s give folks a few minutes. We’ll start in like a minute or two. But yeah, pump for this. [Blake Emal]: Blake and I were- It’s so exciting because I don’t know anybody. [Michael Houck]: Yeah, none of us know anybody. Do you want to do an intro? No, it’s okay. If you want to, go for it. [UNKNOWN_SPEAKER]: Well, I can start. [Sarah Parr]: Hi, I’m Sarah. I’m working on Sitch currently. That’s where the X content’s coming in. [Sarah Parr]: And I recently worked on real world and I did newsletters for them. [Andrew Safranko]: Nice. For a second, I thought you were talking about the MTV show. [UNKNOWN_SPEAKER]: I was like, wait, what? [Andrew Safranko]: Real world road rules challenge, 2005. [UNKNOWN_SPEAKER]: I’m Andrew. [Andrew Cheng]: I’m the new content office manager. So I’m going to be I’m working with all of you on, I’m mostly doing reactions right now, and I’ll be moving on to some other things as well. So nice to meet you all. [Andrew Safranko]: I went next, just because I was talking. I’m Andrew as well. I feel like, am I like one of the OGs? I remember there was like three clients or four clients when I joined you. [Michael Houck]: At this point, you basically are an OG. It’s crazy to think about, because it’s only been like, I don’t know, less than a year. Yeah, but you are an OG at this point. You actually, out of everyone on this call, except for Blake and Ben, you’ve been working with us the longest. [Andrew Safranko]: Crazy. Wow, okay. I mainly do, this one was interesting to me. I don’t do a lot of Twitter. In fact, I don’t write anyone’s Twitter. I do a couple linked in but different writing as we’re all work, but I manage a bunch of the newsletters Heavenly health is one Kayla Barnes venture scaler attorney intelligence There’s more than I love writing socials, but social is always a good thing. So anyway, that’s a bit about me I can go next. [Tanyo Goshev]: Hi guys, I’m Tanyo and similar to Andrew, I mostly write the newsletters for Playground Post, Offlabel and Advanced Genie. And I have been writing some on X and stuff, but it’s like I’m more over LinkedIn type of guy, so I’m curious to learn from you guys. [Michael Houck]: And then, Neeraj, you want to wrap us up? [Neeraj Chouhan]: Yeah, for sure. So hey, everyone. I’m Neeraj. I currently am in Bangalore, India. I’m running a founder’s house here from last one year. And I’m working with David from Workmate on his LinkedIn newsletter. [Michael Houck]: Hey, everyone. I’m Hauk. Good to meet you guys. Okay, so without further ado, let’s dive in. And just for context, the way this is gonna work, Blake will probably drive for most of it, and I’ll sort of chime in in different areas with my take as well. And if you guys have any questions at all sort of along the way or things you think we should cover in more detail, please just like interrupt us, throw it out there. This is totally non-formal, non-structured really. We will have time for like specific Q&A at the end, If there’s something like in your mind that’s hyper relevant to what we’re talking about right now, please just interrupt us and throw it out there. So that’s my spiel. Great. Blake, you want to share your screen and then walk us through it? [Blake Emal]: Yeah, let’s do it. [Blake Emal]: This will just be out of an ocean doc, guys. So it’s not going to be that crazy exciting, but we’re going to make it work. The message is more important than the medium, so it’s fine. [Michael Houck]: All right. Cool. [Blake Emal]: Let’s just focus in here and I’ll enlarge that for those like me who are visually challenged. Okay. So there are a lot of ways that we could approach talking about X, Twitter, whichever you want to call it is fine by me. Boil it down to three core things. We want to talk about the algorithm. We want to talk about hooks and then want to spend some time on Formatting as well all the way at the bottom here. So we’ll talk about those three things If you only learn one of them probably make it hooks But we’re gonna talk about a few things here. And again, if you have questions drop them in the chat interrupt me That’s totally cool. Want to dive into some helpful stuff here If we could just start with kind of level setting on the algorithm a little bit I think that’s a good place to start off and lay the foundation and If you think about how most algorithms work on social, let’s take LinkedIn, for example, usually what happens is you share a post and that is shared to a test group of, let’s say, for example, 10 accounts that the algorithm thinks that they might like your post or they might be within your target market or whatever, however they decide. So it’ll go out to those 10 people and then if they get some kind of signal across those 10 people, This post works for them or resonates with them Then maybe they 10x the next group and push it on to another level of 100 people And they keep doing that until they feel like the algorithm feels like it’s capped off and there’s no more Exponential growth possible and that’s where you’ll see a post die. So on linkedin you might share a post about Writing a newsletter and it resonates with a small group of 10 people really well in the sense that maybe six of them actually Spend a tiny bit of time on that post One of them likes it. For a group of 10 people, that’s actually pretty good. So then it gets boosted out to 100 people and you see similar results. But then maybe it gets to a group of 50,000 people. So let’s say that’s the overall impressions you’re reaching that level. And we start seeing, oh, that group, that’s kind of the cap. There’s not enough market to serve like this content to those people and have them care about it. We’re not gonna put it beyond that. So then the post will die. That’s the general basis of how most algorithms work. And to an extent, X works a similar way, although now there’s more of an added layer of TikTokification to X. Probably two, three years ago, what was working was not exactly the same as it is now. Your follower count mattered more than it does now. And so there are pros and cons to both of these things, right? If I have 95,000 followers, almost means nothing anymore. That’s also kind of cool because somebody with 500 followers can have a post that totally goes bananas and grow overnight. So it’s kind of democratized that part of it. It’s also made it harder in the sense that if you do grow, it’s not a clear indicator that you’re just gonna keep having great posts just because you’re posting. Whereas that is how it used to be. So the way that we mitigate some of that is through a few levers that we know for sure which are listed here as priorities, dwell time, sim clusters, and reciprocity. So I want to dive through those. But before I do that, any questions, anything you want to talk through algorithm-wise or platform-wise before we dive into specifics? [Andrew Safranko]: You’ve got some sort of science behind this, obviously. How did you experiment? Was it literally posting and then sort of tagging yourself internally, like, OK, this is this type of post? Literally play with the algorithm and like, like, get to the point where like, okay, I think I’m doing something to figure this stuff out. [Blake Emal]: Well, some of this is publicly available. So you have to take that with a grain of salt x did open source the algorithm a while ago, it seems like they haven’t updated it in a while. And they also weren’t necessarily under an obligation to follow that open source thing at any point. So what they shared could have just been like, for appearances more than anything. But So a part of this is publicly available information. Another part is trial and error, figuring out what actually works, analyzing other accounts, posting a ton. So thankfully, I’ve been posting on it for such a long time that there’s some muscle memory that it definitely skews me towards these things mattering a ton. There are more things that matter that may be smaller than these things. [Michael Houck]: I think one thing I would flag here is that what Blake is defining is basically the system of how the algorithm works, when he talks about those signals that inform whether a post should be shown to that next level or whether it should die at a current level, that those signals might change. So as you mentioned, a couple of years ago, the signals weren’t as heavily focused on dwell time on a post. Whereas now, that’s probably the number one or number two factor, more so than any type of specific engagement or anything like that. So while those signals might change, the framework that they exist within, in terms of how posts are shown to more people or not, has stayed relatively the same. [UNKNOWN_SPEAKER]: Got it. [Blake Emal]: Yeah, good question. [Blake Emal]: Anything else anyone wants to chat through before we dive into these? [UNKNOWN_SPEAKER]: Cool. [UNKNOWN_SPEAKER]: OK. [Blake Emal]: So again, to reiterate, the For You page is kind of the big thing that has changed in the past couple of years that has made it a lot different than it used to be, for better or for worse. There are a lot of people benefiting from it. A lot of people that complain about reach being really hard now. So you’re gonna see both sides of that. And that’s the reason why, because it’s gone toward basically being text TikTok. So you’re gonna see on your feed, if you don’t pair out your own algorithm, you can pay attention to what succeeds. It’s a lot of these pop culture accounts, meme accounts, small stuff, and a lot of stuff that actually is gonna look like it directly conflicts with these priorities that we’re gonna lay out here. Um, but that’s that’s like broad viral When you’re looking at broad viral, you’re you’re gonna see puns and memes and Like silly things or pop culture things that you don’t even subscribe to at all But you’re also going to see sprinkled in there deeper content from either people you follow or people really related to Your topics or who you follow and the stuff that pierces through there Follows these and that’s the kind of content that actually builds your audience and builds goodwill and probably makes money. Whereas those other kinds of content may go viral. An example of this, I have posted myself multiple times this year, stuff that’s gone viral and it did absolutely nothing. It was totally pointless. It got probably a couple million impressions across three different posts that went viral for me and didn’t do anything because it wasn’t really deep content. It was broader. So you have Cognizant of that we want to optimize for content that actually Hits a goal specifically for a client. So whether that’s more revenue more leads better public opinion of them credibility Whatever we want to gear can work toward that and so priority number one dwell time and profile clicks Really quick dwell time is essentially how much somebody’s how much time somebody spends reading your stuff so There are signals out there that suggest if you can get somebody to read your post on X for two plus minutes, that’s 20 times more, 22 times more important than a like, I had been used to optimizing toward likes and comments on like LinkedIn, for example, and a few years ago was really pushing toward that on Twitter as well. That’s not really the case as much anymore. So when you think through dwell time, I’m gonna open this up to everybody. If I’m thinking through, that optimizes for dwell time, what does that post need to look like to be able to do something like that? Great hook, right? Yes, absolutely. Across every single principle, the hook has to be right no matter what. So that’s why that’s kind of in a league of its own. So absolutely right. Hook has to be great. What other things could contribute to a longer reading or dwell time? [Michael Houck]: Think about the format of the post. [Grace Lemire]: I feel like for my clients, specifically data driven stuff or like images, charts, graphs, people have to read and like understand. Totally. Yeah. [Blake Emal]: So on the visual side, there are visuals that just look really nice. And then there are visuals that kind of require a little bit of processing. Those are actually the better images, usually, because you’re sharing data, a chart, a screenshot essay, whatever, and it actually requires extra time spent looking at it. So if we’re sharing memes, that’s totally cool, by the way. You don’t have to have every single post hit all of these things. We can do memes and stuff like that. But a meme is not going to have the dwell time that a screenshot essay that’s really well thought out or an infographic or some kind of visual like that. So visuals, anything else that comes to mind? [Michael Houck]: Sorry, let me just double-click. On that for a second. So yeah, memes compete on engagement and like conversations in the comments and likes and all that. Whereas other types of posts, these like screenshot essays, and I think what Blake is getting at, I’m going to ruin the surprise here, I think we’re getting at videos, right? Video clips, things like that. Those have more dwell time and they’re competing on dwell time. So it’s like different ways of trying to get, different ways of trying to get your posts to show up in more people’s feeds. Because remember, your post is competing in a bid auction against every other post to show up as the most likely to have dwell time or have engagement from each user. So in order to win that auction, your post needs to compete on dwell time or compete on engagement and win one of those two categories. [Blake Emal]: Yep, exactly. So video, we can just keep going through. I mean, video clips are going to automatically help there. And then another thing that kind of is obvious, but longer content, longer text content, right? Like not every post has to be an essay, but if we, in theory, write longer content, we have a better chance of hooking them and keeping them on that post for a longer period of time, which then has very high signal. So those are just a few things, but throughout this, I want you to think critically, especially in your next batches as you’re writing stuff. Even if you’re not writing for X, you can practice this stuff, but think critically, would I dwell on this thing? Would it require me to take a couple minutes to really understand it? Would I be into it for two minutes? Am I hooked enough for that? What would it take for this post to be more worthy of spending multiple minutes on? So that’s filter one. And then within this bucket as well is profile clicks, which have a pretty equally strong signal. About 25 times stronger than a like, if somebody sees your post and clicks your profile from it, you’re instantly going to basically bump up into that next group, let’s say, where if you get somebody to click on your profile from that initial group of 10 people that see your post, because of the strength of the signal, you’re instantly possibly even going to jump a level in terms of the next group that sees that. So that’s like a force multiplier when you’re looking at, all right, can I click Something that’s worth dwelling on and does it is it so good that it drives them to my profile now similarly if I were to ask What would get somebody to look at a piece of content and click to my profile? What do you think? Drives an action like that besides just saying visit my profile because I obviously you could do that. It might not work it might What does the content need to do to actually make them want to visit your profile and see more? It needs to present a problem and then maybe you can offer a solution. [UNKNOWN_SPEAKER]: Yeah. [UNKNOWN_SPEAKER]: Okay. [Andrew Cheng]: So problem solution. I like it. [Blake Emal]: I think a hot take maybe. [Sarah Parr]: So people would say oh who is this person? [Blake Emal]: Why do they think that? I like that too. Yeah, so we’re looking at like specific formats here and I think if you’re Boiling it down to its foundation A lot of profile visit is about credibility Um, I guess that that’s probably like the core Trust signals that I’d be looking for is does this establish me as an authority in some way to where? If they if they’re looking for something they might think I’m a solution for that, or they might trust me and just want to see more of me. So the more credible our content can come across, the more likely it is to probably drive toward a profile click. And I think Grace had mentioned this earlier, data is a really good way to do that. If we have unique data, unique insights, one thing that I, I’ll probably come back to this point a few times during this chat, but I really like the idea of using uncommon words. And what I mean by that is, Instead of the same hooks everyone’s using the same ways of describing things tons of transition content tons of filler content What if there was just a really high value to word ratio in my content? What if it didn’t read like everything else? Especially within the body because you have more room to work with uncommon language in the body So like you can say something that’s kind of weird and it’s not as punitive but eventually you get to a point even with your hooks where, oh, that’s a Blake hook, or a HALC hook, or a Sarah hook. I know that. If you can become less common in the way that you write, that also can establish that credibility. So data, unique writing, unique insight, yeah. [Michael Houck]: Let me jump in there, because I think that that is getting more and more important, and it has over the past year or two, and it’s only going to continue to do so as AI. As more people are using AI to create their content, and there’s just a higher volume of similar content out there as a result, being able to carve out an authentic voice, an authentic feeling. Like, I don’t know if you guys are in the startup Twitter world at all, but there’s this guy, Mark Lou, who I think does a really good job. And I think Blake even has an example of his further down at this this talk of just, you know, a post is from him when you see it, even without necessarily reading the handle at the top. And one other point I would add to this, too, is that When you’re talking about credibility, another way to do it is literally just to shield the person’s credibility. This was really good for me when I was building my X audience, you know, years ago, was just like, literally, I would say in my tweets, like, hey, I did this, like I raised X million dollars from this investor, or I hired people from these universities, and here’s how I did it, or things like that. So just talking about that personal experience, but in a way, that really makes the person seem like an authority, even if they’re newer on their career, if they’ve done impressive things, let their audience know what those impressive things are time and time again. [Blake Emal]: Yeah, and to double click on that, people literally forget everything you and your clients say all the time. So a big fear you’re gonna have, or your clients are gonna have, and a way to respond to that when they say, well, haven’t I already said that? Or aren’t we being redundant, repetitive? It’s like, yeah, of course we are, we would be failing if we didn’t become repetitive. Nobody notices when you repeat stuff. Nobody remembers, like at this point within X, when we’re talking about the algorithm, 99% of your audience might not even see your posts. You might be getting served posts to like people that are not at all following you because of the way it works now. And only a small percentage see them. And of that small percentage, how many of them are just thinking about your client’s account all day every day? Literally none of them. So, there’s almost no risk to coming across as repetitive. And if anything, like I used to say, in Olden language and Olden English, the way that they used to emphasize things, they didn’t have the exclamation point. What they did was they repeated the word or like the emotion that they wanted to get across multiple times in the same sentence. And that’s how you’re supposed to know the amount of inflection they were putting into it, the amount of exclamation the importance of it, right? So the importance of your client’s topics and the things they need to be saying is directly proportional to how many times you say it. So don’t be afraid to be repetitive. Last thing here, and let’s do this a little bit quicker so we can get to hooks. Forget likes compared to curiosity. Curiosity is a way better thing to be optimizing for. I used to, like I said, I used to go for likes. I used to try to create content that I just think thought people would like it. Now it just doesn’t matter as much. If you can get people curious with a hook, and then really engaging in depth content that increases dwell time, you’re going to be in a really good spot. And then you add that add that layer of credibility to get people clicking onto the profile. That’s where we start seeing our content goals actually getting hit on as opposed to just putting stuff out that might perform okay, but does nothing. So we’re trying to get a away from that a little bit more. All right, this one should be quick. Let’s talk about SIM clusters. So we talked about how the algorithm works at a base level where you keep moving up these levels until the post is deemed dead. There’s also a cluster that happens within your niche, which is something that I have been noticing more and more and more. So to give you an example of what happens here, my account is focused on marketing, content, copywriting, things Every once in a while, I tweet about the NBA or something I’m interested in. And every single time I’ve ever done it, I have literally gotten not just bad reach, like almost zero reach. Every time I do something like that, X just immediately knows to kill the post and not even let me have a chance, even if it’s a good sports take or whatever. So even the quality of the content there matters a lot less. Once you start veering outside of your niche cluster, Now, I have no insight as to how X really, how good of a job they really do at putting you in the right niche and who belongs in that and what that looks like officially. But there definitely appears to be a general topic niche that you belong to based on what you post about most often. And when you veer outside of that, not only does it hurt your reach on that post, it actually hurts your future reach when you go back into your niche post. So the more that you just stay on topic the better off you’re gonna be Any questions on that because I think we can move past that one pretty quick But it’s it’s super important like I had been tempted about this a lot when I first started Some advice that you get is just to throw everything out there and see what sticks But that advice does work if you’re seeing what sticks within your niche But if you’re posting about ten different topics, it’s a lot harder any questions Any questions on that before we move on? [Tanyo Goshev]: How do you, let’s say you want to transition from, you know, want to go into a more specific niche. How do you do that? Or do you just take the hit like in the short term, uh, in terms of reach and then it gets back up since the algorithm knows that you’re now posting about email, for example, instead of just content. [Michael Houck]: I mean, Tanyo, you, you just, you would just need to sort of refine the information diet and what you’re engaging with on the account. So this is the same principle that works on TikTok or Instagram. When you’re setting up a new account on one of these platforms, and X is now included in this bucket, you would literally just like, follow the right people engage with the right content, you’re sending signals to the algorithm, that this is what you’re interested in. And therefore, that’s where you should be clustered. So on X, it’s the same thing, if you’re trying to change the nature or niche down, just unfollow the people who are outside of that that new niche, stop engaging, stop looking at their content, and then start posting about it and replying also to things in that niche too. So there is going to be an initial hit where it’s still trying to figure out like, oh, is this person changing what they’re talking about? But the faster you can make those adjustments, the faster it’ll work. And to Blake’s point, you no longer want to be the broad guy necessarily, because there probably already is a broad guy. You want to start small and then hope that that niche will grow so that you are the guy in a growing niche and therefore your audience grows. This is the same way you start a business. [Blake Emal]: You start with a small target market and then see it grow. Yeah. And other option is you start a new account. Right. Those are the options. Yeah. Okay. Let’s talk reciprocity really quick. This one is the least earth shattering of the three. Everybody knows this one, but just to maybe put some algorithmic explanation behind this, We tell our clients and ourselves, we need to engage because we know it’s important. We know that that’s going to build relationships with other people and it’s going to maybe get us some visibility on a bigger accounts, audience, whatever. The part that gets overlooked a little bit is there is actual algorithmic benefit to engaging with other accounts. Like the more that you think of it this way, if you ran a social media platform, the number one thing you’re optimizing for is how people, you can spend can spend as much of their day as possible on that platform and not being lazy users. So the more that you are a lazy user, the less likely it is that x is going to love you. Whereas if you are a pretty dang active user of x, the chances are that they’re going to be pretty active in promoting you as well because it’s mutually beneficial. They want your engagement, they want everything you do to do really, really well. They can’t do that if they don’t have some assurances that your content’s going to work, that you’re going to show up for other people. So engagement is actually something algorithmically that you should be doing on a daily basis, your clients should be doing on a daily basis. Easier said than done. I know that’s a big pain point. [Michael Houck]: I think for the scope of this call, we can move forward to hooks just because most of the people on this call are on the content writing side and therefore they’re not going to be the ones like doing the engagement. Andrew and Ben obviously will, but they already, they already sort of know this stuff and are working on it. [Blake Emal]: So let’s, let’s move on to hooks. Cool. Okay. That was algorithm. There’s more than it goes in the algorithm, but hooks, hooks are the most important thing we’re going to talk about today, probably because they’re basically 90% of the success of a post. If you write this beautiful thing, but the hook sucks and you failed to hook somebody in right away and grab their attention, there is no second They might as well not exist. It’s pointless. So we need to treat every piece of content we’re writing as if it’s make or break on the hook, because it pretty much is in most cases. Sometimes you’ll still get some people through, but it’s going to kill a ton of the possibility of that post. And the core reason for this is people scan more than they read online. Even when they’re going to a platform where it’s text-based and they know they’re going to read, they still scan for stuff to see if it’s even worth their attention. So your hook is the fastest way for them to still just be in scan mode and actually feel like they want to invest time in it. Treat it like a headline on the landing page, treat it like the hook of a video, whatever. Same principle across the board. It’s kind of, it’s the 80-20 of content as I put there. [Michael Houck]: It’s the, it’s the subject line of your email. They don’t see the email unless the subject line gets them to click. And like for TikTok, they don’t watch the video. Unless the first one to two seconds is entertaining, right? That same principle applies to the first word, to the first sentence, to the first tweet in like a thread, if you’re gonna do it as a thread. [Blake Emal]: Yep. So a lot of this, we kind of know it, hopefully, but if you don’t, this should be good. Please feel free to speak up if you have any thoughts as well. Thinking of it just from a human behavior, psych behavior level, what makes hooks work? Number one is curiosity. So again, optimizing for curiosity across the board, but especially in our hooks. Suspense can work. Another way of saying suspense, I guess, urgency, whatever emotion you’re playing on there. Boldness, contrarian claims work. Specifics work, as Hauk had mentioned, shilling your accomplishments. Basically establishing credibility right away with impressive numbers or data or point of view or experience. Like if I say, I used to work for Mark Zuckerberg, Well, that’s pretty immediately interesting to my target audience because it’s a reputable person that I can attach myself to. So every hook, I think Sean Puri teaches this, every hook that you write should be attached to an emotion that you want to optimize for. And it’s not always the same emotion for every hook. Even within, let’s say we’re writing for one client that is like a venture capitalist. Yeah, maybe we want to do a curiosity hook or one that is just going to get people like, wow, that was really cool. But then another emotion might be FOMO. Another emotion might be laughing. Another emotion might be we want them to feel like, you don’t really want to make people feel sad, usually. That can work, but I don’t think that’s going to apply to a lot of our clients. Start from emotion. I don’t know if we need to go through every single one of these things here, but that’s the core principle of it. I guess this last one polarity, I can double click on a little bit because it is polarizing. Controversy is not the favorite thing for a lot of clients to do. There are tactful ways to do this, but it works because no matter what side of the controversy or polarity you’re on, you are in the conversation. [Michael Houck]: A really good example of this, and sorry Blake, let me just jump in here. A really good example of this is we have a client who has some pretty unique sort of founder habits. And they’re very busy, and they like to talk about that. And so one of the posts that we did for them was we wrote a long form. This was on LinkedIn, but the same would apply. Actually, we pushed it on Twitter, too, I think. But the post was about how they show up five minutes late to every meeting just to avoid the small talk at the beginning of the meetings, and therefore make the 25 minutes more productive and make the person on the other end more urgently trying to make use of the time. And they say it makes more productive meetings. They obviously got a lot of people who are like, you’re an asshole, like you’re not valuing other people’s time, which is a fair point. But this is a strong POV that this guy had. And he wanted to be polarizing and it worked. It’s one of the most engaged and viewed posts on his account. And he didn’t get like, you know, any negative real impacts from it. He was able to grow his following as a result. [Blake Emal]: Yeah. And for each client, it’s gonna be a little different. Some are gonna actually have hot takes some clients they’re gonna say like I would say more often not clients are gonna say oh I’ve got hot takes and they’re they’re so vanilla and that’s okay like we don’t have to invent polarity if that’s not where we can create intrigue but then we need to find it in other ways like are we gonna create intrigue and curiosity by being really smart and again going back to how do we say uncommon things how do we add deeper insights or if we’re if we’re thinking that our PR is maybe too obvious or cliche, how do we conjugate that and twist that to be fully unique? The last thing that we want is to just repurpose what other people do just because it works without it actually applying to our client. So if we see a hook that works, great. But then beyond that, how do we twist that and make sure that it goes that level deeper? Because now, especially with AI, it’s a lot easier to create templates and work off of a template If we can write away from that style We will stand out in the feeds a little bit more and that’ll benefit us and then inevitably That’s gonna change and we’re gonna have to adapt and find a different way thankfully The hook side of stuff there are always new hooks to try out but the psychology behind it is is always the same Can we make people curious? Can we get them to the next line? It’s a it’s the slippery slope that writers talk about if we can just do a good enough job of making them interested right away, and then the second sentence can kind of capture more of that momentum, can we make it almost impossible for that momentum to stop? That’s always the goal. So when I say hook, also it’s worth mentioning, it’s not just the first line always, sometimes it is. Sometimes it takes one, two, three, even four lines to get into the hook for a longer piece. As long as it’s interesting and it’s building on something and it’s not redundant, that’s actually okay. So I definitely don’t want to say this is the core of a hook. This is what it always looks like, Cause that’s just not reality. Yeah. [Michael Houck]: And like, you should be thinking about this everywhere. Like I know that for some clients, for example, we’ll write their bios and things like that. Like I’ll even use the example of my LinkedIn account about section here as, as a good example of a hook, it’s a two line hook. So the about section says, I didn’t know what a startup was until after college. I wasn’t lucky with connections or mentors. And then the second line says, but within a few years, my startup raised 23 million from Andrew and Horowitz and other investors. And that’s all it says above the fold. They have to click the see more button to sort of see how I was able to do that. There’s curiosity there, there’s credibility there. And that’s after they’ve sort of seen the top of the profile that says that I write about content for founders. So they’re like, oh, okay, this is actual relevant journey. So wherever you’re writing it, we should think in hooks. So as Blake said not necessarily just one line hooks Yeah, a couple questions as a framework. [Blake Emal]: You can ask yourself. These are not listed here, but Ask yourself. Is this too obvious? That one’s helped me a ton Is it is this obvious if so got to change it got to take it a level two levels three level three levels deeper Another one again. Is this too common? Am I just saying the exact same thing as everybody else am I using words like the of, for, but, like the really common English words all the time, or does my hook just look visually different? There’s kind of an art to that as well, making your hook look different. And then obviously, would I read the next line is a helpful way to cap off that framework. So a couple of questions you can ask. Usually a hook does have similar elements, so a trigger, something that gets things started, instantly generates curiosity. A lot of times it’s polarity. Or something really smart or credibility then we create tension so you can think of this as like the Pain poke at the pain and then try to create some kind of relief. That’s a very similar copywriting um phrase or framework so creating tension zoning in on the hook Actually poking at the pain a little bit more making it even deeper going a level deeper and then release it make it clear what you’re going to talk about, or what they’re going to get from it. This is only a framework that can be helpful. Not every hook is going to look like this. But if you are struggling to get started, try to look for a trigger, tension, and a release in your hook as a way to get started writing pretty decent hooks. And then you can start finding out other formats that maybe cut out one of these things, and they’re still OK, and you like them. Nothing’s going to replace you. Your own swipe file of stuff that you like and playing around with that. [Michael Houck]: I dropped a link to a Read.AI post that has the Sean Puri power writing course, where he talks about how to engineer certain emotions and how to target certain emotions in your readers. So worth checking that out, for sure. Sweet. [Blake Emal]: Do you want to go through this part, or do you want to just? [Michael Houck]: We don’t have time. Maybe very, very quickly just outline them, and then we can go down into text. Actually like the more tactical stuff. [Blake Emal]: OK, so these are just a couple examples. So if you want to work off one of these to get started, a curiosity hook usually what’s going to make people curious is numbers or unique point of view, interesting knowledge they’ve never seen before, or some kind of big promise that later gets fulfilled in the content. Those are a few triggers that can spark a curiosity hook. If you’re looking to go more polar, again, provocative stance, usually going to work better on either trending stuff or things that are top of mind. So if you’re pulling something out of thin air from a long time ago and trying to be polarizing about it, I mean, good luck, that could work, but usually not. Basically, what you’re optimizing for is can I, no matter what I say, ensure that I’m getting either instant agreement or instant disagreement? In the comments. When you’re working from polarity, what you’re really optimizing for is comments a lot of the time. You’re sparking discussion, you’re getting people to have a take, and that’s the natural place to do that since there’s not a dislike button. Whereas with a curiosity hook, that could kind of go all over the place, could lead to more comments, could just lead to high number of shares. And then a story hook, this is one where if you’re looking to maybe establish credibility and get more profile clicks, This is a pretty good formula for that, where you’re sharing a unique story that only you or your client could say, it’s vulnerable, maybe you have some twists in there, there’s some personal stakes, and all of that likely leads toward people liking you as a person and wanting to check out your profile. All three of those work for different reasons, but I think it’s interesting to note how they kind of optimize toward different metrics. So if you’ve got a certain goal for a certain client, these things can accomplish like very, very different purposes, and all of them are valuable in their own way. There’s a ton more hook types that we could go through, but let’s stop there and get a little bit more in the weeds of it. I’ve got examples on formatting here, but I don’t know if you want to walk through like… [Michael Houck]: Yeah, let’s do it. [Michael Houck]: Yeah, so, okay, so just for the next, let’s say five minutes, so it’s 1.40 my time right now, so at 45 after, let’s come back. But let’s spend the next five minutes. If everyone can go find sort of one post from an account that they really, really think does a good job, and think that that post in particular does a good job of sort of these hook and algo optimization comments that Blake’s been making. We’ll probably go through one or two as a group together. So just take the next five minutes to go find one, and then we’ll call on one or two people randomly from the there. [Andrew Cheng]: you Sorry, I missed this because my dog was barking. Can it be from either X or LinkedIn? [UNKNOWN_SPEAKER]: For the purposes of this, please X. [Michael Houck]: you you Okay, let’s come back. If you don’t have one, just take a pause and we can talk later, but you should be able to find one by now. So, okay, let’s call on one person to get things started. Jacob, what do you find? [Jacob Shipley]: Yeah, I am a big Chris Kerner fan. I’ve been a Chris Kerner fan since before it was cool. We connected a couple years ago actually and it’s been crazy to watch him just like He’s always been big on Twitter, but he’s kind of exploded But he had one yesterday that was I spent 18 months breaking chat GPT 100 different ways big prompts bad tone zero traction But after all that trial and error, I found a formula and then the CTA to click Morris Here are seven chat GPT hacks to unlock a real leverage and save me hundreds of hours I won’t waste your time. So yeah, so that was dope. That’s cool. Yeah. [Michael Houck]: Kind of like niche credibility at the top. And also he talks a lot about, you know, using operator and he’s done some cool stuff with that. So you kind of know that he’s, he’s actually broken chat GPT and done a lot of stuff with it. So there’s that. And then sort of the promise of like, I’m going to help you do it. Right. [Blake Emal]: And then, and then, Exactly what you’re gonna get at the end and the CTA to sort of click in and read the next Read the next post One cool thing about that as well If if any of you were writing on X Twitter five years ago hooks used to have to be like short Mostly and now the the formatting is a little bit more generous as long as what you write is good so Chris is a good example of if it’s good all the way through and It’s okay if the hook is a bit longer. [Michael Houck]: Cool. [UNKNOWN_SPEAKER]: Um, let’s do one more. [Michael Houck]: Um, Neeraj, what did you, what’d you find? [Neeraj Chouhan]: Yeah, I got this Greg Eisenberg’s tweet. I think I’ll just share my screen. [Michael Houck]: Yeah, go for it. [Michael Houck]: Go for it. [Neeraj Chouhan]: So this one, I feel like it works. Uh, I’ve been hanging out with founders under 22 lately living in SF and NYC. I feel this is interesting because SF has certain kind of 22 years old and New York has certain kind of 20 year olds. And then the second line improves upon that my observation of these young founder getting rich with AI. So I feel like that’s a great hook where, uh, He basically starts with, we know that, OK, he’s hanging out with some of the smartest people. At the same time, in this post, he’s making observation that I’m also hanging out with the youngest founders in both SF and VC. And this is how they are getting rich with AI. [Michael Houck]: Yeah, I think he’s also done it. I mean, Grace is the master of writing trends in terms of content and capturing them through content. But this one, I’m sure a lot of you guys have seen the post. From these like you know even 17 in some cases year old kids who are building multi-million dollar cash flowing apps with AI like how AI was built by a 17 year old and now does like millions in revenue and there’s a big story on Twitter so there’s sort of this trend of the young generation even like you know borderline Gen Z at this point who are sort of thinking about life and approaching job of saying, I’ve actually been in the weeds with these people, even though I’m some old mid-30s guy myself, and capturing that. So it’s great. OK, cool. If we didn’t cover yours, please drop it in the Megaphone Studio channel on Slack right now. It’ll take one minute, so you can do that. And then we’ll move forward with some of the examples that Blake identified. [Blake Emal]: All right, couldn’t share. [Blake Emal]: Wrong tab, all right. [UNKNOWN_SPEAKER]: Hopefully everybody got a chance to do that. [Blake Emal]: I’m gonna go through, we got 10 minutes left here. A lot of this still applies to hooks. So we’re gonna chat formatting, but I wanna just call that out that think of it as a lot of hook writing still. I wanna talk about velocity style writing really quick. This is not a big fancy thing. You need to really worry about the the naming of but what we’re really talking about here is Condensing your sentences down to a high value to word ratio and what that usually looks like is sharper sentences That get higher engagement and readability. So we’re looking for shorter sentences where possible I did just mention though that there is more grace in the formatting here So it doesn’t always have to be this but if you want to build the heck what usually works best. Short, simple sentences. You can think of writing in like a fifth, sixth grade reading level. Kill off commas and semicolons if you can. Unfortunately, avoiding the em dash now that it’s been tainted by chat. I’m a big em dash guy too. So that one stinks, but probably avoid it just because there’s scrutiny around it now. And we don’t want to deal with that. Thinking more broadly about this, I just encourage you all to think critically about what are the things when I think chat GPT content or not that it’s inherently bad, but just that it’s noticeable, right? What are those things that I probably want to avoid in my content or my client’s content because I want to have it be less common. I want to come across as more insightful, more unique. There are a lot of different things that you could patch into a prompt to try to get chat GPT away from its original tendencies, those are the kind of things you probably want to write, or, you know, things that are not toward that vein of the same sentence structures over and over that ChadGBT uses, overly fluffy, tons of rhetorical questions, things like that. Instead, if you look at this example, again, Mark Luvion is one that I have seen for a while now, got a couple examples. It is so easy to skim this post and and get information without having to read it line by line. Like I just look at it and I have information bestowed upon me with very little effort. So that’s a formatting thing that if it were written out in full block text would be a lot harder. Now we can go the opposite way though and say, all right, if every single post we do is a bullet list, that’s also not great because bullet lists only make things look easier Sometimes like in an example like this where it’s clearly a listicle post. It’s like a step-by-step just do this And but then it also has some text around it if we’re just posting bullet lists all the time That format is going to become really really tired for people and we don’t want to be doing that either So it’s it’s a balance of plain text and small formatting And when I’m saying formatting I’m usually just talking more about is it easily skimmable? Are we breaking up the sentences? So that I can read it better? Is the sentence structure simple enough, short enough? Are we using way too much grammar, way too many em dashes, commas, stuff like that? But then also sometimes formatting can be a list. Just use that sparingly. Don’t go crazy with the bullet lists on the formatting side. If we’re getting beyond maybe seven, eight bullets, it’s probably an indicator that maybe we need to create a visual for this instead of just listing out 30 things in a post. So sometimes bullet points are an indicator that you actually should make that a visual post and not a written post. One thought per line. Again, this example does a great job. I spend most of my time behind a computer, but I think I found my perfect fitness routine. That’s easy, cost me zero, da-da-da-da. Here are six rules I follow every day. One line per thing, super easy to skim. Again, we want low friction, high retention with as little effort as possible. Avoid filler language vague language. So every word in here pretty much has a distinct purpose in your 20s experiment in your 30s build in your 40s own now Justin could have Expanded on each of these points with a full sentence or more after each of these in your 20s Experiment with XYZ in your 30s build this and this and this thing in this order like you could expand on all that the simplicity of it makes the hook way more intriguing and then if I don’t I don’t think he actually expands on it a ton beyond this, but if this were just a hook and then he was diving into a full thread, it’d have way more intrigue and we could always cover that information in greater depth in the body copy. But the way that it’s formatted is crazy simple to grasp. So that’s usually the go-to for me. [Michael Houck]: Yeah, go ahead. It’s also important to know sort of like audience expectations of each account, right? So each client has a certain audience. Those are different types of people. To and like different types of content. Justin’s audience, they know that he is the guy who is not going to write the big threads. He’s the guy who’s going to be super concise, post two to three of these one post short little tweets per day. And they’re going to be these higher altitude, but still very specific and useful information. He’s not the guy who’s going to go into detail about how he does it necessarily in terms of how he’s experimenting or whatever. He’s the guy who’s going to give you and the guidebook. And if you want to go deeper, you buy his products or you read his newsletter and things like that. So you have to understand sort of how and where each client sits in their audience’s mind. [Blake Emal]: Sure. And they’re going to be clients that prefer to have a little bit longer writing style. They prefer not doing the line break on every line and that’s okay. Like Greg Eisenberg doesn’t do that. You’ll see Grace has normal paragraphs and that’s okay too. A part of that though is that Grace used to do a lot of these things and built his audience and now he has the grace to write exactly how he wants. So there’s a balance there. Like if you just start off writing humongous paragraphs, the chances are it’s less likely you’ll grow than if you follow these principles. But if you do grow your audience alongside these people that care about what you’re writing, you have a lot more grace to fit the format to what you want to do. Here’s an example from Grace here. Uh, again, he, this, he’s got examples that are much more blocky than this. This is actually, this is following like one line per, so not a great example of that, but, uh, does break things up. It’s highly scannable. It’s not a bullet list though. So again, if it were just bullet lists, if they were there, they probably got transformed into some kind of chart or turned into plain text. Uh, so this is kind of the flip side of organize your information visually. But using just text only that’s kind of the default I would say for X or for LinkedIn if you can Format stuff with text only and still make it look visually easy to read Make it look like there’s a high value to word ratio. Like when I’m looking here I’m just scamming and I’m seeing stuff scanning and things seeing money middle-class engineer github entry-level You know like space I’m just saying like unique words that stand out to me. The more we can sprinkle in those unique words that helps visually as well. And then he used a visual instead of a full bullet list or whatever. There are a lot of levels to it, but you’ll get better at kind of seeing that. And then George Mac has a screenshot essay here that we can basically look at the screenshot and also see the way he’s formatted that even within an image follows very clean formatting. It’s not Chaotic it’s it’s on purpose. It’s all fully intentional We have very little time left. [Michael Houck]: So I’m gonna stop on that if that works for you all Yeah, I think we’ll share this doc so you can read through these list of don’ts Afterwards if you have any Anything you want to chat about on that just let us know but I want to open up the last few minutes and I can go over if anyone wants to but no obligation to stay But yeah, I think let’s let’s open it up. So if anyone has questions sort of on what we talked about or things about X in general that we should go over, just flag it. [Tanyo Goshev]: I’m curious about the formats in a sense that I know that for LinkedIn, like the most popular ones are right now like videos, carousels, these types of stuff. So what type of formats work here? Is it threads? Is it long form? Is it screenshot essays in terms of effectiveness? Could you rank them in some order, like right now? It’s hard to rank them because there are examples of each that go mega viral, right? [Michael Houck]: But when you just think about it through the sort of heuristic that Blake laid out at the beginning of what is going to lead to the most dwell time, generally speaking, a great video clip with good text above it will probably probably go the biggest. And we had an account that we didn’t write for, but that we helped them partner with creators. And they went from zero to 20,000 followers in a few weeks because they sort of pioneered this video clip with some write up text above it format. That really does the best, I think. But screenshot essays also do very, very well. Threads can still do well if the hook is really strong also. And I want to flag just before anyone jumps off. We also have the hook library in Notion. So if you guys haven’t explored that yet, please do. There are, I think, literally hundreds of examples in there, tagged by creator, where you can click in and sort of see their hooks. And then we also have the viral examples newsletter that goes out twice a week. So on Thursdays, new templates, and on Mondays, sort of a tip that Blake just writes up. I would highly recommend subscribing to that at viralexamples.co if you have not already. But yeah, Tanyo, to answer your question, it’s definitely more visual content. Threads, I think, are less likely to be the dominant force that they were a few years ago, but they can still work also. Got it. [Grace Lemire]: I have a question that I feel like I don’t know. [Grace Lemire]: This is information I should have known a very long time ago. But so when we’re writing the content for these clients Do can we make posts that are not threads? [Michael Houck]: Yeah, so some clients will will prefer threads and we’ll just run with that for them But other clients will want us to experiment and instead of say three threads per week They’ll say oh I want, you know, five to seven short to medium form posts per week They’re just like one tweet that’s you know, you’re up to 280 characters or if it’s like a longer form medium form post a little bit longer than that. So it’s up to client preference, but we can also push them in a certain direction if we think it’ll, if we think a screenshot essay or a couple video clips will perform better than a thread, we can make that recommendation to them. [Grace Lemire]: Okay. I was, I thought we had to do like three threads per week and I feel like they’re getting kind of just, especially with my client, he’s not controversial in the slightest bit. I’ve tried so hard to like pry some controversial information. There’s nothing and I feel like a lot of the the content and the topics just around like budgeting is just so Bland, but I feel like these threads. [Michael Houck]: It’s just like You should you should you should encourage him to To experiment and like you can even just propose some some ideas So if he’s responding to your questions on Monday With you know things that you think could look good as an infographic or turn it into a screenshot essay instead of a thread just work with Celine and have her have her put that together. [UNKNOWN_SPEAKER]: Okay, thank you. [UNKNOWN_SPEAKER]: I have a dumb question. [Andrew Safranko]: What do you guys define as like short form and long form? Like I’m always like battling myself with like what that would look like, [Andrew Safranko]: So on Twitter they make it pretty easy because short form is 280 characters or less. [Michael Houck]: Basically if you you’re over 280 then you have to click into that see more button to see the full post. We consider short form. Anything that’s over 280, if it’s like slightly over, then probably like a medium form, right? But if it’s like significantly over like that Greg Eisenberg example, either Greg Eisenberg example that both Neeraj or Blake showed, both of those would be long form. And then how does that translate to LinkedIn? [Andrew Safranko]: I find LinkedIn is also like a crap sheet when it comes to long form and short form. [Michael Houck]: Yeah, you know, we should probably do a separate one of these on LinkedIn because the algo is a little different. So maybe we’ll do that Blake next month or something. We can do it on LinkedIn. [Andrew Safranko]: Yeah, I feel like it’s a different world, right? When you talk short form for LinkedIn versus short form for Twitter, it kind of feels different. [Michael Houck]: Yeah, because LinkedIn, you have like only X amount of lines before it cuts off and it sets weights. Than a short form Twitter. [UNKNOWN_SPEAKER]: So yeah, cool. [UNKNOWN_SPEAKER]: Any other questions? [Michael Houck]: Yeah, I have one. [Neeraj Chouhan]: But I think Justin Welsh and the Markler example that we saw, these were very simplistic posts and very, for lack of a better word, massy, which sometimes come around as, like, OK, this is becoming too simplistic, or this is becoming too thread boy sort of content. Any thoughts on that, how to go around that? [Michael Houck]: Yeah, I mean, if you get pushback from a client on a format that’s working for somebody else, just send them the posts by the other person and show them the numbers. Like, hey, this post got a million impressions, or this post got 1,000 likes, or whatever, right? So we need to establish with clients that we know better than them in a lot of ways. And because we do, right? Like, for example, one of our clients had a big launch, I think it was like either last week or the week before. And Tanyo and I kind of co wrote a big launch announcement thread for them and took what they had and made it 10x better. And then they put all this budget behind the launch, but they used the original thread that they wrote instead of the one that we can prove for them. And the launch kind of fell on its face, and they didn’t get the performance they were You know, we were able to sort of be like, well, it probably would have done better if you used what we wrote for you. So we need to establish with them that, Hey, we do have the credibility. We D we are the experts there. And a great way to do that in this case that you’re talking about Neeraj, just literally show them, Hey, this format works. And, you know, maybe your account isn’t just in Welsh’s. Maybe you don’t have the crazy fan base that he does of people who will just consume anything that he puts out, but That doesn’t mean that the content is bad. It just means that by doing content like him, you know you can build an audience that way and establish credibility that way. [UNKNOWN_SPEAKER]: Got it. [Michael Houck]: Thank you so much for the session. I’ll drop off. See you. Any other questions, or should we wrap? OK. Cool. Well, if anyone has anything else that comes up, please just shoot and or Blake DM, or just even open up a thread in the Megaphone Studio channel so everyone can see it. That might actually be better. And I will share this recording as well as the Notion doc that we walked through once the recording processes later this afternoon. So you’ll have that in Slack and we’ll pin it so you can access it at any time. But I think this was great. So maybe we’ll do another one on LinkedIn or on newsletters in the next month or two. [Sarah Parr]: This was super helpful. Thank you guys so much. [Michael Houck]: Yeah. Super helpful, man. Appreciate it. Of course. Happy to. All right, guys. I’ll see you. [Tanyo Goshev]: Bye.