How to Have Interesting Ideas (The Ben Thompson Playbook)

The most important article you write is the second article someone reads, and I do think that volume or quantity is underrated. So that's like 50 or 60 books worth of writing over the last decade. That is an insane amount of volume. It would be hard to have the degree of success that I've had if it weren't for the just day in, day out sort of grinding.

Should we do the day in the life Grand reveal now? So I have a framework, an overall view of the world and how it works. I'm a big believer in hacks and life in general, so the most important thing for success on the Internet is... Ben Thompson is like the OG of subscription newsletters. He makes millions of dollars a year from his writing. If you're thinking about, "Hey, how often should I write? What is my business model going to be? What is online writing all about? And how does the internet work?"—well, this podcast is like a 101 classic on that, and I got to say Ben is a real visionary. His ideas, his way of thinking about the world as frameworks, they're going to give you a sense for not just how the world exists now but where the world is going, and it's going to help you to think about how to thrive as a Creator over the coming decades.

You've got to walk me through a day in the life because presumably you're waking up, you're thinking, "What am I going to say today?" And then by the end of your day you have an article shipped. Yeah, I mean, I think one of the interesting questions that I will get a lot is, "How long does it take you to write an article or write an update?" Well, first off, that distinction is important. So I write two kinds of pieces. I write what I call an article, which goes on the front page of Cheery—that's free for everyone to read—and then I write updates, which are behind the pay wall, and they generally have different formats and they're written for different audiences.

I think the articles are more challenging for a couple of reasons. Number one, I have no idea who's reading them and what base knowledge they have, and they can go viral—they can be read by a million people—and they might be reading the story for the first time or they might not know that much about technology. And so my approach that shifted over time, as you know, the relative base of my readers changes. But like, you have to explain more things on one hand, and on the other hand, it needs to have a hook—it needs to be interesting. But when I started, there were, you know, a number of people that sort of imitated, you know, or wanted to do the same sort of thing, and I thought they were all very smart. They had good insights—there were people who I've been following on Twitter, I'd been reading their blogs or whatever—and they had good insights about technology, and the vast majority of them, uh, kind of flamed out and never really got popular.

And I'm like, "Well, they have good insights," and the key thing that I sort of realized is the difference between my writing and their writing. You know, let's assume the insights were equal. Number one, I had a lot more volume; we should definitely get to volume because I think that's actually really important. But also, my best pieces told a story, and people love stories, and they were pieces that were analysis, but they had a beginning, they had a path, and they had sort of a resolution and conclusion of like, "What did it come to?" What's an example of that from one of your pieces? I was talking about the Disney Charter thing last year, where there's that big sort of standoff, and I opened with the story of the first commercial satellite and the rocket launch and all that sort of thing. To me, that was a great example of what is a killer article for Cheery—where it's like, "You've walked in, like, 'This is really interesting. I've never heard this story before!'" And then suddenly you're like into arcane sort of business details about, about, about carriage disputes. And so that—that was, I think, Ben's different share for Cheery throughout is—to the extent Cheery is really sort of killing it—I am delivering articles that are not just sort of good analysis, but are good stories, and that are interesting and compelling to read.

Now that takes a lot of work, so the answer of "How long do you spend?"—well, some of these articles, it's hours. And a lot of it is literally just figuring out the opening. Like once I can get through the first few paragraphs, I know what I want to say, but like, how do you actually get into it and sort of make it interesting? And some of the less interesting articles, which I think have very good analysis, they have sort of a weak opening. Whatever—I see, you know, just because I, I couldn't—I couldn't get it, but I needed to publish it.

So tell me this: Is the open about you, the writer, like you need to find your flow, and it's almost like going down a ski slope—it's like once you find it, you're good? Or is it for the reader that you need to actually hook the reader in? Well, there are a few aspects to that. So, number one, when I'm writing it is absolutely about flow state for sure. So, you know, I actually think about the way I write as being very similar to the way a programmer is coding: You have the entire structure of the application you're working on in your head, so you know all the pieces and how they work together. And the act of programming is, is substantiating what is in your head, right? Realizing it; that's right.

And so the reason why flow state is really important in sort of getting locked in is because you need to be holding that in your head while you're actually typing stuff—sort of on, you know, on your computer—and looking stuff up and doing whatever. And so, you know, it took a few years of communication and understanding—and frankly realizing I need to work in a different place from my family—in order to really internalize with my wife, which is like, "Look, no, when I'm locked in I..." And she would get annoyed like, "You be...find me working late, I'd come to dinner and I'd just be like catatonic, and I'm just like shoving food in my mouth and then going back." And you're like, "I need an office." She's like, "You need an office; get out of here." Um, so I do work back at home now, in part because my kids are older and they're doing their own thing and stuff like that. But for a while—especially, my kids were young—uh, I definitely reached a point where I was like, "No, I need to actually rent a different place, and I need to be somewhere else just because it's, at the end of the day, it's my job; it's not necessarily their responsibility—they feel like Dad's around, they can talk to him." And so that was sort of an imposition of, of—even though technically I can work from anywhere, you know, there is an aspect of, "I need to—I am in a different place." I'm in a different place mentally, but it's still sort of a different place, and that needed to be matched with being in a different place, sort of physically. Getting in that state is for sure hard, and that's every day; that's hard.

So a huge part of once I actually get started with the writing, then it happens pretty quickly. But actually getting started is hard, and it's a million times harder with the articles because if I don't have that hook up front, then I feel very challenged to start. And I think, you know, just knowing that for me there's still a weight that comes with this—is on the front page. This is going to be sort of broadly available; it's almost like a psychological barrier of, like, "I need a good hook," but the distinction between the articles and the updates is I don't do that for the updates. The updates are, they are very timely—they're usually about something that's happened very recently. So now it's earnings season now, so I'm writing about earnings. I know exactly what I'm going to write about, and the opening is I'm going to pull an excerpt from usually the Wall Street Journal or Bloomberg, uh, to, uh, publications that write in a very predictable way. They have the, you know, it's the traditional journalistic—you have the nut up front and then you have sort of the additional thing at the bottom. I just want the nut up front, like, and that's my launching-off point to talk about that. And so that just makes it way easier because I know what I'm, I know how to get started.

Number two, I know who my audience is. They're people who are interested in what I have to write because they're paying to get it, and so that is just very psychologically freeing in a way. I assume way more knowledge on the audience; I'll just like talk about things without having to feel like I have to introduce it to other folks. That said, the best updates that I write are the ones that do have a common thread—that there is, like, I tile something in part one, and then part two and three sort of brings it around and ties some stuff together. Like, you know, I had something in mind this past week about, you know, uh, I had already written about Google's earnings, but they had a lot of interesting stuff this quarter. So I didn't get into some of their comments about infrastructure spend—which I wanted, or CAPEX—which I wanted to get into, but then Microsoft was all about the, cack stuff, so I was able to bring in the Google piece and then re-tie together how the Microsoft one started. And that was, you know, both satisfying, and I also—you want to leave the reader with a, you know, when they finish they're like, "Ah, yes," and it's like sort of brings it together. And that finishing moment, when they finish reading it, they feel that was a productive use of 10 minutes, like, and they feel sort of satisfied. I don't think—and I don't think it's a conscious thing that they're saying—but if they're leaving saying, like, "Well, whatever," and they go to the next thing, I think you build up sort of a, a well of not disappointment but boredom or whatever. Whereas if you get that little bit of aha at the end, then it's like, "That was great," and then, and it just sort of leaves a positive, you know, sort of affinity for the product.

And so, well, I don't obsess to the same degree with the updates as I do the articles about having that thread. You don't want to not have that thread too frequently. So I did some of the back-of-the-napkin math today—something like 2,480 articles over the last 10 years for your daily updates, public ones. I said you're probably writing 250 a year. I think it's not quite that many; I think it's around 170 or 200, but whatever—it is a lot. Yeah, I've written like multiple books at this rate, right? Well, even then you do five or six books a year worth of content, so that's like 50 or 60 books worth of writing over the last decade. We just need to get that out—that is an insane amount of volume so people think I'm very productive. And this probably gets to your, you know, to circle back to the day-in-the-life thing. Uh, on an objective basis, yes, I am extremely productive. I tend to think on a personal, self-reflective basis: I'm a massive procrastinator who wants to just, you know, lay around and not do anything and like read Twitter, and so there's a bit of a hack. I'm a big believer in hacks and life in general.

It drives my daughter up the wall. I'm constantly like setting alarms for everything—there's anything I need to do, it's an alarm. "Oh, that sounds so annoying." Yeah, I know, 'cause my, my, my phone's just always going off. Uh, but the reality is, you know, to go back to this flow state idea, I am very much a serial thinker. Like, I'm thinking about—like, I'm walking in, I'm not having side conversations. I'm not doing sort of other things. And so my hack in terms of productivity is, I have a daily deadline every single day, and you know, I go back to—I remember being in college and writing a, a college thesis. And so I did a lot of work for it, but the actual thought of, like, sitting down and writing this was, uh, very overwhelming, and it was due, you know, you know, in May when we were sort of set to graduate. And so, basically, you know, I'm not—I don't advocate this as far as being a good writer; I basically sat down like 36 hours straight—a sort of combination of tons of caffeine and then a few beers to take the buzz off when I was like getting shaky, and then back to caffeine. I just wrote the whole thing, and, um, and then I took the whole thing, and then me and some other folks—we were having sort of a graduation party, which we had invited my senior thesis adviser too—and I walked up to him in the middle of the party with like my thesis, gave it to him, and then, like, took, like, slammed a beer—like, right there, um, but, uh, but all this to say is that, uh, for me the hack is deadlines. They make me very productive, and so I've never written an actual book, and I'm fairly terrified to do so because the idea of it being fairly open-ended as far as time—you just need to do a little bit every day. Uh, I, I actually question whether I could actually do that. Huh.

And so what I do with Cheery, on one hand, I think the blog style—and I still call Cheery "blog"—is very conducive to my thought process, and also, you know, the fact I'm covering stuff that's in real time. Like the idea of a book being frozen in time is something that I'm not sure I would like very much. But there's also a bit where it just sort of suits me personally and helps me get stuff done. You've been doing this for what, 12 years now? Uh, 11 years. Eleven years. Do you dream and sleep differently in terms of thinking about your articles now? If I feel I wrote something that wasn't very good, I don't sleep very well, um, and that's the downside of the daily deadline. There's days at work where you're just like, "I don't have it today; I'm not feeling good;" might check out early, and that's just not an option. I'm going to publish every day, and the day that I write something that I don't think is great and I sleep well at night is the day I shouldn't be doing this anymore.

Anyhow, when it comes to, to—so the question of how often it takes, right? Update—the actual physical action—updates are easier than articles. They happen more quickly. But the reason it's hard to answer is the reality is, is I'm thinking about what I write about all the time. And the reason I'm able to be so productive, to your point, is people have a perception of my writing is that I sit down, what I'm going to write about, "Okay, I'm going to write about this; I'm going to do a bunch of research; I'm going to think about what I want to say; and then I'm going to sort of come up with something and I'm going to write it." And that's not sustainable. You wouldn't be able to write however many thousand, uh, pieces it is. The way I think about it is I have a framework, an overall view of the world and how it works, and it's like it's almost like a machine. And so when a piece of news happens, I feed it in the machine and out pops sort of like a conclusion. Like, if A happened, then B, then C, then D, then E. And sometimes it's very straightforward. Like, some news happens, I understand exactly why it happened. All I need to do is sit down and write down the steps in the machine—that's sort of the framework in my head. Sometimes it's a little surprising—that's often where you can make like small-scale prediction, like clearly something else is going to happen.

Uh, CU, there's a missing piece here. Those are the ones, I think, that you have pretty good track record on those sorts of things. What's an example of that? There's going to be another announcement, uh, about, you know, X support for, for feature Y, or something like that, like this doesn't make sense in the context of their business model—whatever it might be. So the third one is, I'm wrong, or there's something—there's a void in the machine. Like, I—this doesn't make, this sort of doesn't make sense. That's usually, I think, where articles do come from; where, "Okay, there's a lot of different pieces here that are being sort of pulled together and sort of make sense, and now I can write a larger—like articles are ideally an articulation, augmentation, or correction of sort of the machine, sort of, sort of, what I've gotten wrong." So, you know, everyone's thinking about, like, you know, AI and CAPEX, and I just mentioned that. So I wrote that piece about Microsoft–Google last week, and it's in my head. There is some sort of article in the formulation stage about the various risks—the relative risk factors of the hyperscalers. Uh, you know, you have Meta—that AI-all for their own product—you have AWS, which is all-by-and-large for third parties, Microsoft and Google in the middle. How does risk sort of align to that spectrum, and what are you doing there? You're looking at the different pieces, and then are you trying to think, "Is there like a unifying framework that bridges this together? Like, how are you beginning now to piece that article together?" Yeah. I mean, I think at this point I definitely have the outline for an article. So then there's a bit about, "Is there some sort of piece of news that is worth being a trigger to write about this?" So one of the things I have found is I will—I've written stuff that's just a pure, like, theoretical articulation of something. It, it usually doesn't resonate that much.

I think that this ties to the story bit. People are interested, again, this goes back to Cheery being a blog, but they're interested in it being tied to something sort of in real life. Sure. So if I were to write a book about aggregation theory or something along those lines, that would be a—here I'm laying out this whole thing. It's meant to be more sort of timeless, given the format that I work in—a lot of my insights—and this is why it almost would be hard if I were to publish like an anthology or something like that. They are anchored in the time and place of which they came, even if the insight is relatively timeless. And so I think there is a bit where I think Cheery rewards the regular reader, 'cause you see the—you see the development of this machine, the evolution, yeah, then Cheery is sort of like a journal of my understanding of the world. And so in this is a case where the format—regular writing every day—and being able to go back and say, "I got this wrong," or change sort of XYZ, and now I think this—in a way sort of writing a book would scare me 'cause it's sort of locked in there. It's like, "Well, no, I don't actually think that anymore," but it's in the book. It's something I enjoy.

Well, you're talking about developing ideas, and you have your doodles, and then also you've mentioned logic three or four times. Are you somebody who is working things out with sketches and a whiteboard, or does this happen in your head, and you kind of get the outline just in the Ben brain? Yeah, it mostly happens in my head. I think the more interesting balance is, how many essays are sort of in my head, fully formed—I'm just writing them out—versus the ones where I vaguely have a sense where I'm going, and by virtue of writing it, it forces me... It forces it into a final form. Sure, I certainly agree with the sentiment, you know, I think Jeff Bezos is famous for, particularly in this, but other folks—where the value of writing is, you have to fill in sort of the gaps between the bullets on the slide, and you have to actually see what you actually think. And sometimes when I'm stuck on an article 'cause I don't have an opening, whatever, I know the answer is I just have to start writing, because the issue is, I don't actually fully know what I think.

And for sure, you know, this gets back to why I still write—even though I do a lot of podcasting these days, and podcasting is definitely easier. I can see, you know, I can see how, like, Bill Simmons— you know, sort of the granddaddy of, I think, Internet writing—even though he was sports—I would view him as a forerunner of, to a certain extent, you know, now you just podcast all the time. It's like, yeah, I could definitely see that. That would be nice. The reason why I do hesitate to ever do that is I, for sure, think that writing makes me a better podcaster because I've been forced to examine—make sure everything that I say is actually correct. I've been forced to actually structure my thoughts in a way that every semicolon is used to its maximum—you know, I'm not repeating myself—but I am sort of building sort of logical things. And I think my goal is—the way I always put it is, I hope the value to my subscribers is not that I'm always right, but rather that I've forced you to think about the issue, and because I've laid out very clearly my thoughts, my assumptions, and the implications of that, you are now equipped to come perhaps to a completely different conclusion. Like if I'm writing about a company and you're the CEO of that company, you have way more information about the situation than I do; but I've now given you a structured way to think about the issue in front of you. You can bring to bear the additional information you have, which may lead to a different outcome. And in that case, just because you disagree with me or didn't follow my advice, I would like to think I've done a large service for you.

So I'm going to throw something out there: You tell me if I'm right or wrong. So when I was scrolling through, on your site you have all the different concepts, and then you can click on one of the concepts, and then there's like three to five articles there, and then you have like a one-line sentence where you've summarized, like, "MKBHD for everything, this is what it is," or "ESPN Bundle, this is what it is." But it sounds like the way that you're talking about your articles is more in outline forms and logic chains rather than one sentence, one-liner that then you're kind of unbundling to see what's inside of that. Yeah, I mean, those sentence one-liners, honestly, get written at the very last second before I publish, 'cause it—something needs to go in that field. Uh, so I do sometimes feel I should give more thought into those and making sure they're sort of a good articulation. To the extent they're good and useful, it's probably an accident—to be, you're not starting with those at all. Those happen right at the end, at the end.

Yeah, I mean, so I have the sort of a broader thought or sense of, and again, this goes back to the sort of machine idea: I'm very rarely writing an article from scratch. Use, like, TSMC as an example. So TSMC—one of the values they provide as a Fab or as a foundry is, you go to them because you want to make a custom chip. So they've enabled this entire world of, "Anyone, you know, all these small companies can have their own sort of chips." But you're not going to actually build an entire chip yourself. What they provide—and this is a big challenge for Intel in sort of competing in this space—is they have like a huge collection of Lego bricks. Like all these different pieces of IP, where you add all the parts of the chip, and then you, as a company, you're adding your special sauce that's like the last sort of 5% or whatever. But there's a ton of stuff about, you know, registers, pipelines, all this kind of stuff that's pretty standardized across chips. And so what you're adding in is sort of the little bit that makes that chip particularly unique and useful for you.

And this is perhaps a stretched analogy, but the way I think about my writing is, I have a lot of the concepts and overall things in place. Something new has happened, and what? How does that tie into what I understand? Or how does that explain how my understanding has changed? The other thing people on Twitter like to make fun of me for, beyond the semicolons, is I love to quote myself. Yes, you do, and it is worth mocking. I sort of feel silly when I'm mocking myself, but there's a lot of times where, um, sometimes it's very, like, pragmatic, where I, I remember—I was writing something about, uh, I think it was like Facebook and their shift to the TikTok-ization or whatever—and my, you know, my thought was, "You know, Facebook—they were constrained by their network, actually—that's actually a function of limitation in how inter-content could be." And you know, the more I write, the more I forget stuff I've written. I went back, and I'd written exactly that, like a couple years previously. And the problem is that because that was when I had the idea and I worked really hard on that article, and I go back and read it, I'm like, "I can't say it any better than this. Like, what do I—what am I going to do? I already made this point a couple years ago." So, number one, it's just like, "Look, I already did a good job writing this; why am I going to kill myself to write it again?" Um, but number two is, again, I think just the overall concept of what Cheery is, as it being a journal of my attempt to understand technology and understand media and understand the internet—the impact of the internet on all these sorts of areas—is it's not about building something new every day; it's evolving. It's, it's, it's touching up the little bits and pieces of, "This is the way I thought the world worked; it's kind of right, but there were sort of a couple things that I missed here." And so, let me remind you of where we're at, and I'm going to sort of add on to that. And I can get that, you know, it probably comes across at times as a little self-mocking, and you know, again, it's just sort of like my signature at this point is that I quote myself. But there it is—actually pretty integral to the way I think about what Cheery is and how that fits into my process, and how it is that I write thousands of articles—because I'm not writing a thousand new short stories a day. And so there's an aspect of, "I've been developing this understanding; there's an aspect of, I've just been doing this every day for so long that the sort of process of how to do it is, is, is…"

It ties into that: How in the world do you edit at the end of the day? Because something that happens to me—something happens to a bunch of writers—is you're in it, and the challenge is when you're in the article you're reading it from your perspective, and you're adding all of these intentions of, "Hey, this is what I meant to say." And I like to sleep on things, because of that, I see it with fresh eyes. You don't have that luxury; like, you're in it. Dare I say, almost in a state of panic, and now you've got to do the edit. So what does that look like? I would say two things. Number one, I will grant the caveat that not everything I write—but some stuff I write—could probably use more editing. And there's sometimes—I, I feel this when I'm going, like, 5,000 words or something that's too long—"This really could be trimmed." And that's just sort of the trade-offs of my model, is that's not sort of happening. The reason I do generally get away with it is I rarely add to my pieces; so, and I think it's, it's—this is a function of, I, it's not just I've been thinking about the article all day, it's that I've been thinking about the article, in a certain sense, for 10 years. And so, ideally, the tightness that I'm going for—where I'm not repeating myself, and it's a very logical thing—is springing from a tightness in my own thinking and a structuredness in my own sort of understanding of the world. And so it comes out in a fairly cogent, sort of state. And so I do, of course, edit, but it's mostly about, uh, typos—mostly about like repeating words, which drives me up the wall. And then, the one actually the one sort of hack that I accidentally developed is now that all my articles also are available sort of as podcasts. I read it, and I always read it out loud before I publish, and that has actually completely killed my typo rate. Like, I have, like, one a month now, whereas I used to have many more than that, 'cause it's like just reading out loud and you can feel that, "Oh, that's, that's sort of stumbling."

I do think, I think I write, by and large, how I talk. You can see I'm talking in long extended, sort of answers and sentences here, but hopefully I'm talking in a way I'm not rambling too much. Or, you know, I'm rambling to a certain extent, but it's not like random rambling and I'm tying stuff together, and ideally I'm linking it back to—it's still in my head. You ask me what my typical day is, which I haven't yet answered, but there's a bit where we're going to come back to that—that's in my head, and that's just generally how I write and, uh, you know, I think about the human brain in terms of, like, computers to a certain extent. Like, I just have a very large amount of sort of like L1 cache, like, uh, you know, where the whole thing is sort of in my head and I can retrieve it quickly without having to go to, like, like-memory or whatever might be.

Do you want to—should we do the day in the life Grand reveal now? Well, I think the challenge, actually, in some—this sort of gets to your editing point—is like, anyone else, you're going to be more high energy, more capable of doing stuff sort of in the morning. The issue is that because I'm in Taiwan, and if I'm doing interviews with people, or I'm just talking to, like, my assistant—that I do all the writing and stuff myself—but like all the taxes and accounting and all that sort of stuff, that's mostly all people in the US. And so my mornings are usually occupied with just admin stuff or doing interviews with people or phone calls if I have calls. And also sports—I’m a big sports fan—and sports that are on at night in the US, or restaurants in the morning in Taiwan. And so if there's a foot basketball game, I'm probably going to watch the basketball game. Um, and so I will wake up very excited; maybe, you know, I don't always know what I'm going to write, but sometimes I know I'm going to write—I'm like, "This is great; I'm going to finish the day early; it's going to be awesome." And then stuff happens during the day, and then I get lunch, and I'm kind of sleepy after lunch, and it takes a while, and I start procrastinating. And then you get to like, 3:00 or whatever, like, "Oh my God, I have to read this update; I totally forgot about it." So one thing I have done is, uh, I've sort of compressed my entire week into three days. And so, because the fourth day that I do now is an interview—and that's usually happens in the first few days of the week—and basically I just, uh, somewhat physically destroy myself for those three days. Then there's a lag in the afternoon where I'm being lazy and procrastinating and not wanting to do it. I'm kind of low energy, and then sort of the panic sets in, and then, you know, the late afternoon, early evening, is, is about sort of getting that written and getting that done. Um, and then I sort of crash out at night, and, and wake up... I didn't quite imagine you with like a quill pen and, you know, some sort of very meditative office writing every single day. But it is striking how much panic is involved; "Oh my goodness, how big of a part of your writing process that is," and frankly it's one of those things that, uh, I'm not particularly proud of it. I wish it was better. Like, I wish I was more disciplined. Was it like that when you were at the Daily Herald in college? No. Same, same sort of thing.

So, yeah, I mean, in that case, you know, and that goes back to one of the things that when I took over the editorials—or the opinion, or the editorials—when I was a senior, the student newspapers would do editorials, and they would do it about, um, random national things, like once a month or something—no one cared. What we—what I wanted to do was like, "Oh, we're actually in a unique position for a very specific set of issues," and so my idea was, "No, we're never going to write about national issues 'cause no one cares; we're just some random student newspaper." We are going to write about these things, and we're going to write an editorial every day, or Monday through Thursday, and then Friday will be like letters to the editor and stuff like that. And we did it, um, which mostly manifested into me doing it every single day and writing like all those editorials. And, of course, they were all—had to do it every day. You had to come up with something; you had to figure something out; and then you—and then you were doing it at the last minute. And I loved it. It was awesome, and, like, and it was successful. Like, I think of our five objectives, we at the end sort of did an inventory, and we felt we had succeeded on four of them.

And so there is a bit where that was sort of a foreshadowing of what I did with Cheery, was, you know, I now, again, do an interview now, but it's been four days a week for a very long time. It used to be five days a week, uh, and I write about technology. I write about media. I'm not—I'm not partisan to the extent I write about politics, like I wrote a little bit about, like, some of the, you know, uh, KL Harris or JD Vance or whatever and their views on this stuff and what impact that might have, but I'm presenting it like, "This is just sort of my view on it." I'm not saying who I prefer or XYZ, 'cause like I—my goal is, of course, I have opinions. My goal, though, is to have an impact on technology in the area that I am focused on; and I think ultimately that can have more impact on the world and more influence by sticking to what I'm good at—just like the student newspaper. Hey, if you're, if you're hammering on something like student government, they care what the student newspaper says. If you write about, like, the Iraq war, no one cares; it, it sort of doesn't matter. And so gauging what you're targeting—what you're focused on—to what you're capable of, and granting some degree of authority to talk and write about it, I think, is something that is useful and important.

How do you think about your own reading? I mean, I remember many years ago you wrote a piece called "I Think It Was Called The Voters Decide," which was inspired by a book called "The Party Decides." And so there's a book there, and then there's a lot that you've built on, Clayton Christensen, and then you probably wake up in the morning and thinking through headlines—like, how do you think about your reading input, and how intentional are you about that? Uh, my reading is extremely intentional, in that I'm not sitting around reading books; but there are days where literally I know what I want to write about. I know there's a specific thing I need to understand better. I think some of the best analysis pieces are historical in age and that they pull into, like, I think a lot of things companies do go back to things that happen at their founding, and that because that is still a part of the culture—and to understand how a company thinks, you have to understand their culture. To understand their culture, you have to understand their founding. And so there are days where I will be writing about something, and I will, in the morning, be an extremely fast reader—fortunately, I will sit down and I'll read through an entire book really. And then I, then that book will be a heavy reference for, you know, that I'm writing and reading about. Now, there are certainly foundational books, like you mentioned, sort of the Clayton Christensen and things on those lines. And once I've read a book, it's sort of in the—you know, it's in the repository that I might sort of pull on, and things on those lines. I mean, certainly, Clayton Christensen is, is the biggest influence in that regard for sure, but yeah. So I read a decent number of books, but not when I'm just like sitting on the chair relaxing. It's like that's part of my workflow: I'm just going to plow through this, and, um, you know, at this point, you know, I know which topics I'm going for and which things I'm sort of really seeking to understand, um, and those are very satisfying ones.

Or there's other ones where I remember there's an article I wrote ages ago when Apple was changing, you know—it's part of that update—sort of explained this and what the implications were and sort of XYZ. And the good thing about that is it's directional, but obviously that's in there sort of in the long run, you know, by the time, you know, the AI stuff came along and Nvidia and all that sort of thing, like, I knew at a very pretty deep level how GPUs worked, and that was arguably born of a, you know, an intense day of research. Because I, you know, I think the biggest compliment I can get is when someone will comment on Twitter or they email me and say, "You know, I have to say, like, you know, for someone that's not in my field, you really got that." And, and, and that's what it—it's great 'cause you want to—we all know, like, the "G-Man Amnesia" thing or whatever, right? You read something, you're like, "Oh, that's really good," and it's about something you write about, like, they don't know what they're talking about. And it's like, I don't want to ever do that, and sometimes I do. And then I value greatly readers that will write in and say, "You got that wrong." And, and the next day, you know, it's not fun, but John Gruber—who writes, "Staring Fireball"—has a saying that I have wholeheartedly adopted, which is he says, "I love to be right, and the best way to always be right is to immediately correct yourself when you're wrong." And I think that is certainly something I try to live by. You know, I'm going to get stuff wrong, especially when you're writing every single day, but I try to not just issue corrections, but also I walk the reader through my thought process—"How did I get to the wrong thing? And you know what was that bit in my chain that was... that was mistaken?" And I think that's, it's not just good hygiene in general. I write pretty authoritatively—just that's my style. It's like, "It's pretty direct—it's, be, because of B, because of C." I'm not going to, you know, like dilly-dally around it, which I think is an important, just general writing tip: You got to go for it. You can't be, like, making excuses or hedging your bets, you know, too much in your writing. If you, you say what you think—but that's also risky because when you're wrong then you kind of look like an idiot. Tell me more about that "going for it." I want to hear more about that.

I state things: "This happened because of this," as opposed to, "Well, one could, you know, um..." And again, I'm not saying I'm perfect in this—I do hedge my bets, you know, and you could probably tell, sometimes there are days where I do leave sort of an out for myself. So I'm not sure, but by and large I think you want to—if you think something is the case, your whole value as an analyst is not being a, you know, not being a stenographer—like, you can just say, "Oh yeah, company said this; therefore it's this." Your whole point is to find, you know, we talked about the PowerPoint to the essay bit before. There's a bit where what companies say are PowerPoints and the role that I can bring is to fill in the space between the PowerPoints of what companies say. And you don’t do that effectively by just listing every possible thing or, on the other hand, just repeating the PowerPoints. You have to actually, the value you provide has to be something that's—this is what's in the middle. And so you say, "What's in the middle?" right? And ideally you're right, uh, and sometimes you're wrong. You have to own that. But there, I think that's the best approach, because if you're sort of mey-mouthed around it then it's not interesting writing. Like, no one finds it sort of compelling.

Yeah, one of the things you do a lot is you'll look at a keynote and you'll say, "This thing happened, they framed it like this." And you'll say, "That's my wedge, and now I'm going to analyze that, and then this is what it means for the cultural grain of Google or Apple." Yeah, I mean, I love keynotes. Um, the—I love the first five minutes of a keynote 'cause to me it's like, they're trying to set a frame for what's to follow. And to me, the frame they're trying to set is more intriguing than the products. Like, why is this the frame that they're sort of looking for? Um, you know, like, Seth Adella does a lot of, like, he tries to tie stuff back in historically, which obviously I try to do as well, so I find that appealing generally. But you know, does that make sense? Is that actually a relevant sort of historical comparison? Like, you know, and what? Why are they doing this? What does it say about the products they're introducing?

I think the risk that I have to be careful of is, first off, there's just a general confirmation bias. My whole bit is, I think I already know how the world works, and I'm plugging this news into it. So I have to be very, very careful that I'm not looking for stuff that fits my pre-existing worldview. The funny thing is that sometimes that makes me get stuff wrong. So I overcompensate, actually. "I had it figured out!" But I'm like, "So eager to look—but that's okay. Those are mistakes I'll accept," because it's a function of my process of making sure I don't give in confirmation bias too much. The other thing is because I do actually think the narratives these companies say are important—because they speak to the underlying focus of the company and its culture and things on those lines—at the end of the day, the actual, like, results do matter, and the actual like specifications do matter. And so I need to make sure I'm still digging into—does the actual, like, numbers match this sort of narrative that's being put forward?

So I want to talk about, I want to talk about introductions and how you think about them, and you're talking about stories mattering. So I'm going to read this to you, and I want to hear your thoughts on it. So this, "From What Clayton Christensen Got Wrong"—"There's no question. This, at the very beginning, there's no question. Clayton Christensen, who developed the theory of disruption, is Silicon Valley's favorite business school professor. For me, diving deep into his thinking in a corporate innovation class was a breath of fresh air for management theory that explained all of corporate America. But for its most successful company, Apple, what's going on there?" That's probably like the origin article for Cheery, even though it was four to five months after I actually started, you know, the bit was true. I, you know, quite—you’re in, you're in sort of business school, and they have all these sort of theories about how things work. I mean, I have a few business school bits that are, you know, somewhat relevant. I mean, number one is the fact I was in business school in the first place. I sort of grew up on the internet, uh, you know, interested in technology and interested in the business side of things. And by growing up on the internet in, you know, the '90s, 2000s—NB were terrible, right? They were, they were worthless. Um, but I sort of grew up in a blue-collar, sort of area. No one really went to work for tech; it never even occurred to me to do that sort of thing—even though I was super into it in sort of the Doom era. And, and so I was in, sort of, the 2000s—I was in Taiwan teaching English—and my wife’s like, "You know, why don’t you, all you do is read about and talk about tech. Why don’t you go work in tech?" I'm like, "That's a good idea. I should work in tech." And I'm like, "Oh, what? I'm like what, 27 years old with, and my, no career-relevant career experience?" Like, the only option to me was like to get an MBA—that was like the way, sort of, like get legitimacy in the US job market. And so I did that. It turned out it was a great fit, like, just 'cause I had a lot of implicit understanding of things like, uh, leverage and margin and, and game theory and all this sort of stuff that suddenly, I now had was given a language to sort of articulate things like that with. And so the MBA was great in that regard. But still, like, a lot of technologist people in technology were fascinated by Apple, very interested in it. And, uh, well, there's actually a step up. There's one more thing. So I go back to business school, first day, first class, you know, management 101 or whatever it's called, and I'm flipping through all the case studies we're going to do for the semester, and there was zero technology companies. And so I go to the professor afterwards, you know, "You know, I'm an annoying student like that." I'm like, "Why are there no technology companies?" And the answer is like, "Well, there are broad principles that you learn from these cases that are applicable to, sort of, you know, companies in all industries." And I'm like, "I don't think that's true." And this is actually one of the core things about Cheery, which is technically it is true, but what I think makes technology unique is the nature of zero marginal costs and zero transactional costs. By zero marginal cost, you know, one more person visiting google.com doesn't cost them anything. Do they have substantial server costs? Of course, but like one extra person doesn't cost anything. And this applies to basically everything on the internet. Zero transaction costs is, Google can serve 7 billion people without a sweat, because their actual interaction is completely mediated by computers. And the implications of this on business models and strategy, I think, are very profound. And it was very, and, and to me it was a major hole in the curriculum—that this wasn't the case. Yes, you had all these equations that you could use to analyze businesses, but if you're putting a zero into an equation, the equation comes out looking totally different because zeros cancel so many things out.

So, number one, from business school is there's an entire arena of strategic analysis that, at, I'm, you know, at a top five business school, is not even aware that's needing to be addressed. Let me just pause you one thing that I think is a really good prompt or thing that you find yourself saying when it's like, "I should write." Is when you're in the midst of an expert—like, you were in business school—and what did you say? "I don't think I agree with that." That, I think, for a lot of people, they say, "I think I'm missing something." But somehow when you find that, can you start asking yourself, "Maybe I'm right. Let me go explore that," because if you find yourself saying that, and you are right, you have the seeds of a really compelling piece.

Well, I mean, there's a certain bit, I think, to be a writer—or to be a writer, particularly of what I do, which is my own site on the internet—that entails some combination of narcissism and delusions of grandeur, um, where I think I'm right, and I think I'm right enough, and compelling enough, that I can just put something up there, and you're going to read it, you're, it's going to be spread around, and it's going to be interesting. Um, I'm—I'm not sure this is good career advice. I'm fortunate that, by and large, it turned out to be the case, you know, I started Cheery with 380 or something followers on Twitter, no one knew who I was, and there was no hand up or favor given or whatever. I just went up there and wrote what I hoped was compelling stuff. There was social media back then; there was a lot of currency to be gained from sharing interesting links, you know, and different than today, when links are suppressed and there's so much stuff out there in, by and large. And so I certainly benefited in that regard, but yeah, there's a bit like that's just my personality and thought process. I'm like, "I'm an English teacher from Taiwan, and I'm going to go tell the strategy professor that I think your curriculum on day one is BS, 'cause you're completely missing, you know, sort of a..." That's awesome. Yeah, sort of. I love that.

So, that was, that was sort of, uh, the first sort of business school bit. But the other thing was Apple, as I mentioned, and Apple did everything wrong, right? They didn't have a division—they had a functional organization—a divisional organization. The way they did marketing, the way they did product development, all this stuff was supposed to be commoditized. And like, why were they able—and at that time, there was a real fever around the idea that Apple was doomed; it was imminent because just like the Mac and PC, uh, the iPhone is going to fade away, and all developers go to Android because it was open—then the iPhone would become a fractional part of the marketplace. And that, that whole argument around that, broadly speaking, really, really ties a lot of Cheery threads together. Number one, everyone gets the historical context of Mac versus Windows wrong. It was not, "Mac was there, and then Windows came along and knocked it off." DOS was there before Mac, and that was downstream from, sort of, IBM, and the DOS market was massive and large before the Mac ever even shipped. So it's actually the inverse, where the iPhone was first, and Android sort of came later. And it's amazing that this is just a basic historical fact that basically everyone in the world does not understand. I greatly respect Professor Christensen, as I mentioned. He's had more influence on me than basically anyone else. He got that fact totally wrong, and it meant his entire thesis fell apart. But that—there's a bit where that's why it's really important to go back and read those books. Go back and make sure I get the bits right, 'cause if you just go along with the popular sort of thing, you actually don't go back and assure that, make sure that what the myth is, is what actually happened. You, you can actually get huge swaths of analysis wrong because your one of your core assumptions is sort of mistaken.

Well, let me talk through a few things that I've taken. So one of the big ones, like I said, is you got to know stuff. Another one is to, once you have that background knowledge, to just trust your point of view, and to not be hesitant. Like, there's something that you're doing where you're taking that point of view, you're like, "Okay, I'm going to go for it." And I think part of that is probably temperamental, where you're saying, "Okay, this is how I am for sure." I mean, like the, um, you know, if you do like personality tests or whatever, I'm like high and disagreeable—0%—I'm 100% sure that's the case. And I think, by the way—and this does—I think, color internet discourse generally: Anyone who's successful on the internet is probably very highly disagreeable, like, 'cause the amount of abuse and disagreement and people coming at you—that you, you know, PE people weren't meant to see that, get the level of feedback that they do, right? And most people just sort of want to get along and want to be along, and so, and you can see them wilt and, and, it's a very difficult thing to deal with. And it's not—and I'm not saying I'm impervious to it, like, there's, you mentioned not sleeping—there's because I haven't slept because, like, something got, like, spun up into a completely different arena of people that I wasn't aware of, and I'm not focused on, and I'm getting all these sort of attacks, and it's, it's very difficult and challenging to sort of deal with. Uh, but at the end of the day, um, I am temperamentally sort of, "I don't care." Well, the other Ben Thompson prompts—so to speak—are, "What are some places where there's an expert and you understand their theory, but it doesn't explain something?" And then, "When's a time in your worldview where you're saying, 'Okay, something isn't lining up here'"? And I resonated with what you said earlier, and just think about, to all the daily updates that I've read, some of the times when you started, you said, "Yesterday, the day before, I got something wrong—here's what I got wrong." These are almost more instructive, and I felt that as a reader, those moments are when I feel you as the writer more intimately than when you're talking about aggregation theory or something sort of coming from on high—you're much more professorial, right?

Well, yeah, you're getting inside my thinking, which I think is, is intrinsically sort of more intimate in that regard.

You told me earlier, "Come back to volume. Let's talk about volume." Well, so you know, this ties into the sort of daily schedule it ties into the fact that I still think of myself as writing a blog, and it ties into the business model which is, you know, as a contrast. Uh, there's a guy named Dylan Patel who writes, called Semi Analysis. Uh, and he, he—and he will get these incredible scoops or details about the next NVIDIA processor or what the production pipelines are or, in, in like components and things along those lines— incredibly valuable stuff, like, market-changing stuff potentially or strategy-changing stuff. And, you know, we've talked about sort of business models and what he has. It's a Substack. And this is a guy; it's an amazing internet success story. He grew up posting on forums and, like, he was like a Reddit moderator at some point, and now he's like, like, the authority about stuff like this, and he charges something like $500 a year, and he only sells annual subscriptions. And the reason is he might only write one article a year that actually matters to you—but you really have to read that article—and, and for him to charge what I charge, $15 a month, you know, uh, that would be weighing so much value on the table, because you just sign up for a month, read the article, and then cancel. And you like, because it's, you know, that what he's selling is these occasional super, incredible, powerful and useful sort of insights. And so he's doing a good job of maximizing his return on that. What I do is I charge $15 a month—it used to be $10—it sort of made its way. I've done two price raises. I just, just did another one. Part of what I am selling, frankly, is the ongoing production of content. And it's almost like it's like monetizing my hack, which is—my hack is I have a daily deadline set. I have to publish every day. What I'm delivering to you is any one piece I write is hopefully insightful. Oftentimes, my most insightful things are in the free articles, which you could just read sort of anyway. But what you get when you sign up for Cheery is you're going to get three a piece on Monday, Tuesday, Wednesday, and hopefully a very compelling interview on Thursday. Obviously, I have a podcast; you get all this, you know, it's a big bundle. Hopefully something's going to be interesting to you. But it's that regularity of delivery, and you're paying not for the archives—you're paying for the confidence—and even if you don't realize what you're paying for, just knowing that you're going to get this in your inbox every day. And one thing that I think is, is kind of interesting is the reason I can do this for so long is I'm writing all stuff that I'm interested in. If I'm like trying to cater to my customers, I think that's a recipe for burnout. The one time I did almost burnout is like 2019-2020. I was writing about, like, regulation and antitrust and Congressional hearings, and I'm like, "Oh, this is actually horrible." Um, like, it's burning me out. I like—I work a lot; that's not what's burning me out. It's writing about stuff that I don't— that is kind of soul-sucking, at least from my perspective. But then, and you also over-index on sports, 'cause that's what you love talking about. I do write. Well, med—I mean, number one is that I'm in the media, so that's interesting. Number two, I do think the media is, uh, it's, it is indicative of everything that's going to happen—what's going to happen to everyone. The canary in the coal mine, right? Because it is digital inherently, and the media—look, the reason why there's like, on Wall Street, TMT (Tech, Media, Telecom) or maybe it's Telecom, Media, Tech—remember, those are all zero marginal cost industries that are the connective sort of tissue. And so it's interesting. I do get the media. I would say my analysis of the media is wrong more often than, is about tech. I think a lot of that is, I don't fully understand all the incentives. Like, I think a lot of my stuff about Hollywood has been directionally correct, but I've gotten details wrong in part because Hollywood's a city of big personalities and, like, people do dumb stuff 'cause there's prestige in it, or sort of XYZ; whereas tech—you're, you have people doing dumb stuff for those reasons, but much less so. It's like tech is pretty, hues, pretty closely to its broader, broader incentives. But yeah, it's also fun to write about all of which is to say, I mean, so some of it is self-serving—it ties into my hack—all this sort of bit. But I do think that volume or quantity is underrated, and this is a massive reveal versus state of preference thing. Everyone will say, "Oh, I just want you to write good stuff. Like, don't worry—when you write it, I'll read it." And again, maybe I'm just telling myself this, but my sense is it would be hard to have the degree of success that I've had if it weren't for the just day in, day out sort of grinding. And, you know, you're going to get something from me—and yes, that occasionally leads to stuff that is maybe not as high quality as it should be; maybe you could use another round of editing. But I suspect that the actual process leads to far more interesting and compelling stuff because there's no choice—you're going to have to write regardless. And I do think it's something people find valuable, and it sort of ties into their, their, their daily life.

I haven't done a lot of deep analytics about my readers, uh, I think it's important to focus on what—just focus on what you're interested in. That's the key thing to doing this for a long time. But I have monthly subscriptions and I have annual subscriptions, and at this point, the vast majority of my subscribers are annual subscribers, which is great, because that means they're committed—they know they want to get this—and for me, it's freeing. Like, if I have one bad update, okay, fine—I have a year to make it up to them before they have to sort of renew. But one thing I did notice that was interesting—I was trying to understand, like, the lifetime value of the different groups—is I had to cut out people who would subscribe in the first month and immediately turn off auto-renew, and those are people, it makes sense—they wanted to try it, or they want to get one article—and so I just separate those from people that, that, that stuck through. But when I did deeper analysis, it turned out my best cohort was people who subscribed and turned off auto-renew and then sometime in that first month they subscribed annually, and they were people that liked that they wanted to try it out, and I always deliver to them every day, and they're in the habit, and at some point they're like, "Yeah, I want to keep getting this." And like, "Why don't—why even bother with monthly?" It's a discount to go sort of annually, and sort of, well, I think, you know, I have a lot of subscribers that have, you know, they've been there for years and years, and they're awesome, you know, I still have my initial set of subscribers that, you know, they got me off the ground, and I'm incredibly grateful to them. And I've had a lot of people that have churned, and, you know, my churn rate is very low. But it still obviously sort of exists. I, you know, I bounce them to the—they still get the free articles sort of bit. And it's funny, like, a Ro company should understand their business. I don't want to understand my business too much, to a certain extent, because as long as the number is, is not going down then, then things are good.

So, I got to ask: What is it like when you're seeing someone come in? It's like... Jeff Bezos or Mark Zuckerberg—like, that's a big name. Yeah, I, I—I’ve never sort of tried to read them, um, and know who's coming in. Actually, there is a funny story about this, which is, you know, currently Cheery runs on my own sort of system—which will be available broadly, sort of, hopefully soon—but for now it's sort of like a custom sort of thing. But for a while before that, I was, I was working with the service Club Memberful, and I was like, "When they first signed up customers, they, they actually watched my first version of Cheery, 'cause this business model didn't really exist." So it's sort of all glued together by myself, and then they started this, like, and they did a great job. But they emailed me one time saying, "Hey, just letting you know, you have three subscribers that we accidentally double charged. We've already refunded them," and then they gave me the names. It was random person A, random person B—top five Fortune 5 CEO was number three—and I'm like, "Wow, I didn't know he was a subscriber." It was like, he was, like, totally random. But I—I don't like knowing. It's weird, actually—it's a real paradox: On one hand I don't like knowing, and I usually forget pretty quickly because I don't ever—I want to write what I think. I don't want to ever write because I know someone else is, sort of, reading whatever might be. And frankly, they don't want that either. Like, I do. You know, most relevant, well-known, sort of executives do read Cheery, and I've had a chance to talk to several of them. The reason they find it valuable is when you're a CEO, everyone you talk to inherently has an incentive to tell you what you want to hear, right? 'Cause just, it's inescapable—even if you have the most disagreeable, straight-talking person ever, there's always going to be some sort of constraint there. And so they like reading me or talking to me because even though I'm lacking a lot of knowledge that they have, because I'm not internal to the company, I'm just reading their financial reports or watching their presentations or listening to their earnings calls, uh, I have no incentive to lie to them or no incentive to, sort of, I just say what I think they can sort of—and it’s interesting, to the extent I will get pushed back from companies about stuff I write. I never get pushed back from CEOs. It's always people lower down—the CH people, marketing people, or even senior VPs or stuff like that—who disagree and say, "You don't get this; this is wrong." I've never had a CEO push back at that level. What's going on there? Because they appreciate getting a different point of view, of course—they want the public record to be in a certain regard, and, and perhaps they will email their head of PR, say, "Email Ben and tell him..." But, um, they—I've never gotten push back from CEOs personally. And my perception—and, you know, again, just to the extent I've talked to them sort of in person—is they're so hungry for a distinct point of view that doesn't care about them that, like, I don't care what they think. Like I'm saying what I think. My business models are not dependent on them. They're paying $15 a month like everybody else, and I don't even know if they're subscribers by and large—and I try not to know. And so the flip side of the paradox is, of course, I'm thrilled they're reading, right? Like, at the end of the day, I want to have an impact. And, you know, I used to work at Microsoft, and, you know, I care. I had opinions about what the company should do, and one should be—and the great irony is I almost certainly have way more impact on them than I ever would have had if I stayed for 10 years and sort of been promoted up the chain and been in the internal, sort of, political fights. And that's amazing. Like, there's a bit where my core audience is a very small collection of, like, CEOs, and everyone else is helping pay, pay the bills, you know, sort of, go along with it.

Well, one of the things that I want to make really clear is this is a compliment that's going to sound like an insult, so bear with me: There is something now sitting here that is very unremarkable about you and Cheery, and it's because the business model that you founded and that you pioneered has now become a big part of the world. And I think of that in two ways: First of all, you—the founders of Substack—have explicitly said that you inspired their model. That's the first thing. And then the other thing is, on any media companies, have really become a thing both in writing, where, you know, I guess you had before—you had bloggers like Andrew Sullivan and Ezra Klein—but now we have people like MKBHD, we have people like Joe Rogan. It is a big thing now that single-person media companies can have all this reach. But you go back 10–15 years ago—that would have been insane. They're going to disrupt ESPN. They're going to disrupt MTV. What are you talking about? Well, I mean, at the end of the day, from the user perspective—time is time. So every minute spent—Rory is a minute not spent reading the New York Times or not spent reading, so from that perspective, uh, I am competing on an equal plane with the largest media companies in the world. And so, and that's all that matters. And then for it to pay off from my perspective, the advantage you have as a single-person media company is your cost structure can be totally different. Like, like my, my cost structure for, like, the first four or five years, I was literally one person. I did everything. I did my taxes. I did my accounting. I answered emails. I did sort of all those things. And while that was busy and stressful—particularly since I wrote more back then than—than I, I do now—then, you know, it was pretty insane, the amount of stuff I was doing that meant my, my only costs were basic credit card processing fees, right? And, and like, you know, 'cause my time was obviously, you know, valuable, but it wasn't—I wasn't paying a salary. It was just sort of like whatever came out of it. And I'm working in a zero marginal cost business—for me, to send an email to 100 people, yes, email costs money, but on the order of, say, cents. It's the exact same as to send it to 1,000 people or to send it to 10,000 people or to send it to 100,000 people. The beauty of media is, uh, it is on the internet—it is perfectly scalable. On the back end, I can write an article that reaches a million people, or reach—or theoretically can reach billions of people. Now, do billions of people read it? No. But sometimes a million people do, um, which is amazing. Media is not scalable on the front end, because you have to keep writing stuff every day. Like, like the—and so that turns people off from it. It makes media, I think, maybe often a poor investment, and that's a real challenge, but that's, that's an opportunity for the people that are capable of that—to have that generative function.

So, is it appropriate for you to start with what your day-to-day looks like? 'Cause that is the part—that that is marginal costs. Like, I have to literally sit down and write that update every day. Once that update is written, then I get the internet advantages—I get to send that to, say, a million people who are subscribers or people on my mailing list or whatever it might be. So that is—and, you know, and I wrote a lot about that when I was getting started and like, "Why this is an opportunity," and the different frames of reference. And so I'm very proud of the model. I'm proud of Substack. I'm proud of everyone that is sort of doing this. And I hope to some extent, it gives—even though I've never been the CEO of Microsoft or Meta or or Apple—hopefully, you get some sort of credence that I know, to some extent, what I'm talking about. Because, like, the fact that—yeah, I was the first to sort of really do this. Again, newsletters existed on Wall Street in particular—those are more of the, uh, you have like 100 subscribers, and then you start charging $10,000 or $100,000 or, sort of, whatever it might be. This idea of, like, you know, Stripe was really the, the core piece—they started in 2009, the product came out 2011, I launched in 2013—and this ability just to take credit cards easily, do a subscription, not have to muck around with PayPal or whatever, and basically, make it up in scale like that. The internet lets everything be made up in scale, and that is a challenge—like it's funny, people will bemoan the fact I only have a thousand listeners on my podcast or I only have 500 people reading my blog. It's like, "Well, let's step back here. Let's go back to the '90s and tell someone you sit down and a thousand people listen to what you have to say every single week." Like, it's, it's pretty remarkable. I think the reason why people don't fully appreciate it is, number one, it's everywhere. Number two, through social media, you see all the success stories. You're not gauging yourself, sort of, appropriately—and that's a problem with all sorts of stuff. And then also, the monetization is obviously hard because you're dealing with zero marginal cost. Content advertising operates at scale, and so the key thing with Cheery is—I mentioned, I'll price compared to Dylan Patel, but I'm very high price compared to most content on the internet. And it's really about maximizing the average revenue per user; which is, if you're selling something, you have to be confident to charge for it. And you know, if you charge $100 a year, you have 10,000 subscribers—that's, you know, 10,000 true fans. You know, the, the classic, sort of, essay is, "If you have $100,000 a year, and that's a pretty good outcome." So when you see people launching a Substack trying to charge for content and it doesn't work—what went wrong? What is the mistake, the strategy mistake, that people repeatedly make? When I started, I was the only one doing this, sure.

So that certainly makes me want to be humble about that fact, which makes a big difference. One of the real big challenges is—and this is where the volume thing, frankly, comes in—is you need a way to acquire customers, but you also need something to sell. Hi, I solve that by just writing a massive amount—some of which was free, and some of which was paid. If you're only generating one content piece a week—one article a week—which is frankly a lot—how are you going to balance customer acquisition with actually having a product that people feel compelled to pay for because they want to get access to it? I think that's really hard. It's, it's—and it's harder today. Just, a big part of the Internet advantage was social media; my best marketing is my customers telling other people, and that was when links were being pushed on social media in the way that now they're just... That's right. And so that makes it astronomically more difficult today.

You know, I’m now able to launch new things like I start a new podcast and I can overnight have five-figure sort of listeners because I'm lending the sort of audience that I gained before sort of into getting that off the ground. So what do I say to someone who's trying to start a new podcast that has no audience and no, whatever? It's—I just want to be humble about that, sure. But also, there's something fundamental to Cheery: That businesses can pay for it, it provides value; WR, right? Write consistently, really good stuff. I mean, the thesis that I have is: "The most important article you write is the second article someone reads." So they go on your site, and they're like, "Wow, that was a good article." And then maybe they leave, and they come back later because they follow another link, or they click somewhere on your site, and they read a different article, and they're like, "Wow, that was a really good article." And suddenly there is a level of consistency and expectation that you've established with that reader. And so some of the things I did with Cheery are, funny, I mean, I joke about it's a "word-of-mouth blog with a name no one can pronounce," um, but it's also memorable. That's why I can't remember the name of that box. "How do you say that again?" Uh, Cheery had a custom font—when custom fonts were a completely new thing on the internet—uh, Cheery had the hand drawings, uh, Cheery was orange. The reason for all that is, is I wanted to capitalize on that second visit where people would show up and they'd say, "I've been on this site before because I was just following links on Twitter," and then, and then, "Oh, I see that drawing. Oh, that’s a guy that does hand drawings," and what I wanted is that user to start to develop an expectation that this site has good stuff. And, and I did that. I started Cheery with the goal of monetizing it, but the way I always thought about it is, when I monetized it, I wasn't going to drop a paywall; I was going to do more stuff, because I wouldn't have to have another job; I could write more. I was probably insanely turned out—okay, but probably insane at the time—confident I could generate more stuff. I'm like, "I have so many ideas." And so when I add the daily updates, like, "Yeah, I still have the free stuff, but now if you want more, you can get more stuff." And I wanted to build a desire and linkage in people that, "Like, this is a place to go to—to get good stuff. When something happens in the world and Ben doesn't write about it, I'm disappointed. Like, I wish Ben would comment on this sort of thing." And then, I, you know, a daily update comes along: "Oh, you want me to comment about news of the day as it happens? $10 a month and you can get that."

So I want to do a little segment here called “How the Internet Works.” So I'm going to give you framework sentences that you've written, and I want you to basically unpack them. And I think what we can do is, we can get the Ben Thompson model of how media works online. First thing is this: This is Exponent—my favorite episode of Exponent. Episode 12—it's called "The Internet Rainforest." And the thesis is this: That the internet enables big winners and small, focused niche players. Yeah, well, there's lots of names for this. It's the barbell effect. I've done the thing called the smile curve, and this is basically the idea that the internet—the implication of zero marginal cost and zero transaction cost—and this is sort of aggregation theory—is, everyone had a vision of the internet of being a million sort of winners because you can go anywhere; you can do anything. The reality is, when you can go anywhere and do anything, you're overwhelmed with abundance. You have too much stuff. And so the pro, whereas in the analog world the controlling factor is scarcity—and so in the analog world, it's who can get you stuff. So it's distribution that matters.

On the internet, anyone can get you stuff 'cause it's all free, uh, and so you have too much. So what matters is discovery—who can help you find the stuff that is sort of interesting. What this means is, the platforms that control discovery—they get more users; they get more users; they get more supply because that's where the users are, which makes their platform better. So they get more users. You get a virtuous cycle. That the implication of the internet is not that you have a lot of winners, but that you have very few, very large winners. And so this is the Googles and the Facebooks of the world. This is all very obvious now, but I think, you know, Cheery, in the early days, was really explaining: "Why are the expectations around these companies wrong?" Like, the great shame is, I had no money when I started Cheery. I would have been much better off had I had the money then, and then just investing in my thesis back then of like, basically, "CU, like the big," the big five completely outperformed VC, completely outperformed anyone over, sort of, the running of Cheery. And I was explaining why that was the case—why are they—why are all these companies, despite the fact they're the largest companies ever, still completely undervalued. And that, you know, was obviously, sort of, turned out to be the case. That's probably why I have a lot of loyal subscribers; 'cause I help them make a lot of money.

So, uh, so good for them. Um, so you have those large players, but at the same time, all these bits about the cheapness make all these small creators completely possible. And the key thing is, is traditional media can't shrink down enough—their cost structures were built for a different world. They're built for a world that's predicated on having white manufacturing operations, which is printing presses and delivery trucks and things along those lines. Which entailed a certain sort of lock-in in their market that they had. If you're online, if you can get just a few subscribers that are really, or a few people that are really loyal to you, and really like your stuff—and there's a gazillion niches in the world—and now you can address the entire internet, so you can pull, I have subscribers from over 100 countries, like, you, you can find the people that like what you want. And then you can charge them again a high average revenue per user, and you can have a, have a very good life. Um, and so that is the case in industry after industry. Like, like, like internet cost structures and niche—and "find your tribe"—versus zero marginal costs, serve everyone, sort of, "get these aggregation effects." Everyone in the middle is screwed.

And, um, that started with print, and then it happened with, with, is happening with the rest of media. And you're seeing it happen in all sorts of industries, right? Let's do a string of sentences now that truly changed my life. I remember reading these in my college dorm room, and I was like, "I think, uh, that might actually influence what I do for a career. Here we go: People underestimate the scale of the Internet. It's not going to appeal to your content. It's not going to appeal to the whole world, but if you're going to—like, if you like it, you're really going to like it." Yeah, this gets into the bit about the whole "world's your addressable market." People, you know, you read, like, every story post: "The addressable market is literally on the order of billions of people." Now, there's the, there's the classic phrase everyone makes fun of: "If we can only capture x% of the market," that's actually true in this case, like—and, you know, it's funny, there's a lot of like, well-known VCs and folks in that area when I laid out the pay version of Cheery; he's like, "You know, love your stuff, but no Cheery, that's not going to work." Not, um, and it's interesting: People whose job is to understand the scale of the internet did not scale the internet in that particular case. Like, it turns out, uh, there is a good number of people that are willing to pay because they really like the content. And, and we see it—it's not just me; there's all the success stories on Substack. There's all the YouTube success stories to your point—these individual creators—and there's no, there's no cap on that. People have, like, subscription fatigue, and they're like, "I think they underestimate the number of niches," and they're always focused on themselves. There's a lot of stuff. And actually, their problem is they're too interested. There's too many things they want to subscribe to, and, and they're frustrated about how much they're spending on subscriptions, without—and they forget about the fact that most people—everyone is different. Everyone is not like them. And people will subscribe to, you know, the most arcane sort of niche focus sort of things, and, uh, and it pays off.

Let's do one more: "Most of what I read is the best there is to read on any given subject. The trash is few and far between and the average is equally rare." Well, I mean, I think this is just an internet effect, which is—and this is a function, this is why discovery matters. What bubbles up, what gets traction—the links that get shared are shared because they're good. And you know, this is a reason why you, too, about things like misinformation or whatever—one of the worst things that exists in terms of intelligent conversation about the internet is search, because you can go on and search, and you can find absolute garbage, because this is, like, the whole TikTok-ization of things. The vast majority of videos created on TikTok are trash; what the magic is, is that if you, if, say, 0.001% are great, if you can ramp up the volume, then the absolute, on the internet, the absolute number matters most. And, and so you just, like, if you have a huge amount of volume, even if a tiny percentage of that is good, the number of good things is quite large. And it can be so large that it starts to overwhelm sort of other sorts of things. One of my favorite lines on this is from a little book by Aubrey Mishenko called "Human Is Media," and it says that before the internet, curation happened before the publication of something—and now, curation happens after something gets publicized. And that's what you're saying: You just let everything go. Everything can be shared, and then you have search or algorithms that then say, "This is the best stuff." Now, that's what we're going to show—that's right. So what the curation is is another word for discovery, which is, which is the, these platforms that help you discover the best stuff—and they are heavily motivated to engender massive amounts of creation. And actually, it is a positive for these large platforms if there is gargantuan levels of junk out there, because the implication is there's lots of stuff being created, which means, you know, there's going to be lots of good stuff too.

I want to talk about writing conclusions. So let's go back to what Clayton Christensen got wrong. Here's how you ended it: "Apple is, and for the last 15 years has been, focused exactly on the blind spot in the theory of low-end disruption—differentiation based on design—which, while it can't be measured, can certainly be felt by consumers who are both buyers and users. It's time for the theory to change." So, as you're thinking about ending a piece, what is really important to you? Well, I think that was a good example of how I could be very wordy and use a lot of punctuation. So thank you for that. Uh, number two, you are calling back to an article I wrote in 2013, which feeds my worst fear, "Is that I peaked early and it's been downhill ever since." Uh, well, it is the staple Ben Thompson piece, uh, but number three: Conclusions are hard, for sure. I think the best conclusions—this is kind of a cliché writing device—but they tie back into the introduction, and they, they, they, they not the threat as it were. And so I started that piece, as you quoted before, saying, "Everyone admires Clayton Christensen. Um, he's very smart, but he's got something wrong." And so at the end I say, "What he got wrong." So that was, it was fairly—yeah, I, and I do think when you talk about—you asked about editing earlier. To the extent I do edit again, I think I write with a pretty high degree of clarity, but what if I do rework something sometimes because it's too long? I just got to take something out. Um, but the other big thing is I'm not pulling the thread tight on the conclusion. And so, I want to—I want to, you know, go back sometimes when I'm writing the conclusion, I will copy the introduction at the top and I'll paste it down at the bottom so it's right there in my line of sight, and so I'm thinking about, like, "How am I tying this together?" And sometimes, like, I can't really tie this together, and then I need to go back into the essay and put in a couple of guideposts so that I'm, I'm, I might, driving sort of to pull it back in. How does that work, usually? When I came up with the introduction, I had a sense of where I was going. Sometimes the piece goes in a different direction, and I have to, like, sometimes scrap the introduction, which is very sad, uh, but sometimes it's like, "Okay, no, it's there, I just—there were a couple guideposts I needed to add so that when I get to the end I can sort of link it back to where I started." Um, so I think those are the best conclusions. Um, conclusions are hard. Though, like, the, the, but I do think they're important, like, you know, you want to—you, you want to, and they can be risky 'cause sometimes you, to get a good conclusion, you have to, like, maybe go a little bit more on a branch than you want to say, sort of, whatever it might be. But I would say the challenges of the conclusion are tied to the challenges of the introduction, and again I think that's sort of rudimentary 101 in a certain regard when you're editing and you're in a story—how do you think about the pacing of that story and what the right kind of tempo is for a reader? It's definitely a challenge between saying everything I want to say, and having it be short enough, compelling, you know. Again, I think that's where, when I could use more editing—often, that's the case, or it's late at night, or I'm just like, you know, like I wrote a piece about, it was getting related to European regulators or whatever, and I had a whole section in there about, uh, you know, using the European internet, and this annoyance that everyone wants—like, I kind of can understand why they're so obsessed with data because everyone's trying to collect data that's totally unnecessary. And I think it was a good aside and a good anecdote. It was also, like, a 6,000-word article that, it wasn't directly pertinent to it, and so I was going through, I'm like, "I really should cut this. This is not necessary." But I was really in love with it, and it was also like 3:00 in the morning and I'm like, "Now, whatever, it's, it's going in there." Um, and so I got—I got a bunch of tweets about the—I need to keep it. I need to keep it, like, "Ben must have had a good vacation, 'cause his first post back is like a 6,500-word tirade against the, uh, the European Union." Um, but, uh, so yeah, there's probably a bit to be, totally honestly, where I've gotten worse about this over time because, like, people are going to read it, and so, like, you're going to hear everything that I have to say. I should be, I should be more disciplined about—it’s a bit also the classic saying, "If I would have had more time, I would have made it shorter." Where, when I'm going through, like, summers are super hard, uh, because actually being in Taiwan is a big benefit, 'cause I work during the day and I publish in the evening—it's morning here. Here I wake up at 3:00 A.M. usually, and I write. And there's a bit where, here you wake up at 3:00 A.M. on days that I'm writing. Jeez. So, did you get your writing done today? Uh, I didn't publish today.

So yeah, in July, in the first part of August, I don't do interviews, so that's sort of like my break for the year. I only publish three days a week. But, um, and so there's a, where, and I still finish later than I want to and I'm under the gun, and I feel like my summer updates get a little long, uh, in part of that is, is honestly just a function of, and I almost hesitate to talk about it. It's not my reader's problem, like, "I'm busy or I'm waking up early." I think it's important for writers to adopt the sense that, at least from my perspective, I'm selling a product. It's, if you want access to this, you have to pay for it. I think—I, I, I'm very—this happens less, but early in the days, people, there's a lot of personal appeals like, "Oh please support me," and a lot of writers—when you write something, you want everyone to read it, so they want stuff to be free. And I talk to a lot of writers like, "No, you're running a business; you need to make, like, you have to give someone a reason to pay." And a reason to pay is not charity, and a reason to pay is not that they want you to be popular. The reason to pay is because they want to read what you write, and you're not letting them read it. And there's a bit where a lot of writers are too hesitant, and they're not disagreeable enough to basically tell their, their potential customers, "No, you're not going to get what I write unless you pay for it." And you need to have a clearly articulated value—"What are you selling?" And, um, and so to that end, I don't like talking about myself. I don't like talking about, "I'm super busy or I'm dealing with XYZ." And so this is, you know, late, or whatever—that's why I don't take personal days. Like, if I'm busy or stuff's going on, that is not my reader's problem. My reader signed up to get an email every day, and my job is to give them that email every day, and my problems are my problems. That said, my updates get long in the summer 'cause I'm busy and waking up early.

So, if I were the chancellor at Northwestern at the business school and I were to invite you back and say, "Hey Ben, you're going to teach a semester, and we're going to train people how to do what you do. What would be on that curriculum? How would you structure it?" I mean, I think the most important thing is what you're going to write about. Like, everyone thinks I have the best job in the world 'cause I get to, you know, quote, "Sit in my pajamas and write about the big five tech companies." But if you wanted to do that, you are literally competing directly with me and directly with Twitter; everyone has comments about, like, "I'm getting paid to do what millions of people do for free on Twitter every day," which provide their opinions on the big five tech companies. So you're jumping into a very big pond in a world where you have no inherent advantages of, like, geography. Like, you could have a bunch of sports commentators in the US before 'cause they were limited to their geographic area of where they're published. Once you're online, you're competing with every sports commentator in the world. Now we're just going to read Bill Simmons. They're not going to read your local commentator or whatever it might be. The dynamics of competition completely change what you can do, and this goes back to the niche bit: You can find your own pond. What you need to do is, the most important thing for succeeding on the Internet is defining who your target audience is. So, for example, there's a guy, Neil Cybart, who writes about Apple, right? He writes about Apple every single day; and I write about Apple. So we're nominally competitive, but if you really care about Apple, like, he's the guy you'd should sign up for; he's going to write about Apple every single day. And that is him choosing, in some extent, a smaller pond because he's only reading about one company. I'm reading about a bunch of companies and media and stuff along those lines, but he is owning that pond. And, and I think there should be someone that writes about Amazon every day. There should be someone that writes about Google every day. Uh, Casey Newton writes about, like, social media networks generally, and he's been very successful sort of doing that. I think it—again, it's a cliché, but I think it applies very much: If you want to be the successful fish in the pond, the key is making your own pond, not trying to sort of defeat the other fish. Have you ever heard of a publication, somebody fail because they chose a niche that was just too small? Probably there are a lot of failures out there. I mean, it's the old playing on the internet, right? With bullet holes in it—like, you see what succeeds and what doesn't. I do think that grind is underrated—like, it's very, it's very easy to get started. You've probably been thinking about it for a long time. You're like a band putting out its first album with songs you've been curating for years. And the great bands are the ones with the great second album, and where there's an ability to generate consistently new and interesting sort of insights and points of view. And it's interesting, because I do think there's a bit where, because I do it daily, I think about it all the time, and so my process of coming up with ideas is always running, to a certain extent, there's a bit where the way to not run out of stuff is to make more stuff. Which is sort of counterintuitive—like, creativity isn't a well that you just borrow from; it's, it's kind of constantly replenishing.

I do think there is something to that. Certainly that gets back to the concepts we talked about earlier about having a view of the world and a system, and you're sort of augmenting and building on that as opposed to trying to build a perfect edifice of like, "This is my castle of technology." No, this is my sort of much more organic, always changing, always evolving, always being different sort of view of how things work.

Lovely stuff. Well, I got to thank you because I don't think this podcast would exist without your work and some of the things we were talking about with the laws of the internet, and it's because of your writing and your ideas that this exists, and thank you very much.

Well, thank you for having me.

Yeah.

Winston Churchill wasn't just the Prime Minister of the United Kingdom—that's what people know him for—but he was also a prolific writer. He wrote a novel, two biographies, memoirs, and, of course, as Prime Minister, speeches. He'd spent roughly an hour working on them for every minute that he spoke. So if he spoke for eight minutes, he'd spent eight hours in prep. And yes, I know he's controversial, but man, there's a lot to learn from his writing. So what I'm going to do is I'm going to play you a short clip from a speech that he gave in 1940, and then we're going to break it down together.

"We shall fight on beaches, we shall fight on the landing grounds, we shall fight in the fields and in the streets—we shall fight in the hills; we shall never surrender."

So let's break this down like a good battle plan. The structure of Churchill's writing here is simple and strategic. A commander-in-chief can get his squadron, his unit, on the same page with repetition, and that's what Churchill is doing here. There's no mistaking the core themes here.

What is he doing? He's using the word "fight" four different times, and then he talks about how the British military will fight in five different places. They're going to fight on the beaches. They're going to fight on the landing grounds, in the fields, in the streets, in the hills. So you see all the buildup there, but the entire paragraph is building up to these words right here: "We shall never surrender." That is the main point at the end—that is the climax that everything builds up to.

Churchill also uses style to get his point across, too. Writers these days, you sit in your fifth-grade English class and you'll be told to only keep what's necessary—cut the fluff, get rid of the excess. But Churchill does the opposite here. You'll notice that rhetorically, the volume and the diversity of places mentioned is actually more important for him than the literal meaning of each place. And Churchill—he could have just added emphasis by adding a bunch more places with the word "fight." So you'll see here he's got "fight, fight, fight, fight." He's got all this "fight," but, you know what he could do? He could just add a whole row's worth of stuff. "We shall fight in the cities, we shall fight in the skies, we shall fight in the forests, we shall fight in the little Italian sandwich shops." I'm just kidding, right? But he could have just added stuff. And look at this: The order of the locations is immaterial. So he could take "fight in the skies" and he could make "beaches" down here; then we'll move it up here, then we'll take this and we'll go over here, and we're sort of like shuffling things around, and this all works.

Here's what matters: All you need is, right at the end, you just need, "We shall never surrender." This just needs to come at the end, and if it does, the whole thing works. Now, why is this? It's because speech writing is different from the kind of writing that you usually get on paper. It's this series of phrases—these little phrases that are serving up to the punch line at the end, right? They're just building, building, building into "We shall never surrender." And you could arrange any of those little phrases; you could take the first one, make it the seventh—rearrange them however you wanted in the paragraph—it would still accomplish its purpose. A sense of timing is important, though. Adding all these little buildup phrases, right here—what are they doing? They're increasing suspense, right? Up until the point that you start losing people's interest. And the more engaged your audience is—like when you're speech writing—the more of these little buildup phrases you can add. So yeah, you could say the majority of what Churchill is saying here is fluff. He could have taken all this and compressed it into one thing; it would have looked like this: "We shall fight everywhere and we shall never surrender." Eight words could have had the same meaning, but that wouldn't have been memorable. We wouldn't be talking about it almost a century later. Instead, Churchill took 31 words, and all of these words, right here, they raised the stakes of what he's saying—they're giving his speech an element of suspense, right? When he wants it the most—in this drumbeat of repetition: "We shall fight, we shall fight, we shall fight, we shall fight"—it's paving the way for his eventual climax, "We shall never surrender."

Well, that was fun. Who knew that arts and crafts class would come in so clutch, huh? Well, look—I publish one of these writing examples every single week on writingexamples.com, and if you go to the site, you enter your email right at the top of the page. I'll email you the latest one whenever it goes live.