Five ways to find something worth writing about

The way my life has shaken out lately, I've found myself dedicating quite a bit of time to thinking about the process of writing for gain: whether I'm writing this blog (which I mostly do because I don't think I'm capable of not writing, but the authority boost and the donations are a nice bonus: as an aside, if anyone would like to help support this blog project, you can do so at the bottom of the page), producing marketing copy for my website, writing LinkedIn posts for promoting stuff even more... an awful lot of the way in which we build status and get what we want in this society and our industry goes through writing of one kind or another.

This led to the realisation that an awful lot of us, one way or another, have been dedicated to talking a lot and trying to cut through all the noise and get attention, all without having anything useful to say. The vast bulk of how we communicate, especially in the kind of communication that happens in these status games, is mostly a question of "who can shout the loudest": a competition that a lot of us simply aren't going to win. There are other ways to play the game though, and this is, I think, what this article is about (it's all a bit weird and given that in some sense this is about content marketing, it's out of keeping with what I usually write, but hopefully it'll be interesting anyway).

All blogged up and nothing worth saying

Every society, in whatever shape, has at least one platform (and often multiple) where elite competition takes place. In the Roman Republic it was the cursus honorum, entry into and participation in the senate and all of the multifarious tasks that an elite Roman was expected to do for his patrons (large parts of it, of course, mediated by military participation). That was where, if you were a somewhat educated and wealthy Roman, you would compete with your peers for social standing and for the best positions in your society. In mediaeval Europe, this competition happened in tournaments and on the field of battle through skill at arms. In Classical Athens, success in rhetoric, oratory, sporting competitions and the writing of plays could have similar effects (though of course, there was once again a significant military element at play).

In our society, elite competition happens largely on the fields of LinkedIn. If you're a well-educated and well-to-do member of our societies who wants to compete for prestigious white-collar jobs (and most of my readers likely fall into that class, even if we sit on the margins of it: however poor and financially strained we are, it remains fundamentally impossible to actually de-educate a person), you do that by being heard. We write short-form content on social media and longer-form content on blogs, we speak at conferences and make videos, we publish podcasts, we do all kinds of things to get people to see our faces, know who we are and believe us to be clever and possessing abundant authority. As access to the elite becomes steadily tighter, we find ourselves under more and more pressure to succeed at these contests.

This has led to some perverse incentives. The goal, after all, is to be seen and heard first, and to communicate something of importance second if at all. You're expected to do this using formats generally associated with the communication of ideas in text, and you're expected to do this in volume. You see that a lot of the people who succeed the most do a lot of this. But there's no explicit demand that you communicate anything truly serious, and in fact a lot of people who succeed the most don't really say anything serious. As a result of this, people feel an awful lot of pressure to speak, to publish and to express, and the fact that a lot of the time they don't have much of a truth or coherent idea to express is irrelevant. Consequently, we're drowning in slop and banality.

You see it everywhere: formulaic, almost scripted-feeling, whole sections of phrasing almost identical to a whole slew of other articles written on similar subjects and in similar places, as though they've only ever read that kind of essay and take that to be what good writing is. LinkedIn posts and infographics that repeat the same vacuous points (the 80/20 rule is a favourite for these) over and over again. Podcasts and videos with amazing production values that are more or less about nothing. There's so much of it, and those of us who want something worth reading have to dig through an increasing amount of it to find the good things. And almost all of it ends up being made because people are competing to find one of the few remaining positions of comfort and security in our society. We're pushing anyone who wants a good position in our society to constantly speak and make noise even when they have nothing to say.

Slop is useless to you

At this point my usual instinct (along with fully half of the fediverse and almost all writers, I suspect) is to bemoan this kind of slop production as polluting the infosphere. It makes sense for us to be upset about that: after all, for us it means that it's increasingly hard to find writing of any substance in amongst the slop, and we find ourselves frustrated by the flat, formulaic and oftentimes plain shitty writing that is, for the most part, what gets published these days. However, because I'm wearing my marketing hat today, I'm not going to do that.

Instead, I'm going to talk about the fact that even if you're a purely mercenary operator with no moral compunctions about producing slop or using it for marketing or whatever, slop is almost certainly useless to you. Here's why. Your goal when writing (unless you're just making art for yourself and your friends, which I'll temporarily exclude from consideration) is to participate in these status competitions, or otherwise to somehow gain something of value to you. You might be writing quite explicitly to make money, you might be writing to establish yourself as an authority in a professional field (this is mostly where I sit, though if people enjoy independent writing, they are of course encouraged to donate to the cause), you might be attempting to get a product you've made or a company you run seen. Given that, you find yourself fighting for the limited attention of your audience, and you have broadly two options for winning that contest.

The way in which slop aims to win the battle for attention is by flooding the zone: the hope is basically that you can put out enough good-enough slop that your content will occupy some non-trivial fraction of the total writing on your choice of subject, meaning that someone searching for information on the subject will probably hit something of yours thanks to sheer saturation. To do that, you need two things: the ability to put out content that passes a minimal bar of quality in real volume, and enough ability to distribute your stuff that you can get eyes on it without it necessarily being very good. For an individual consultant or a small company, neither of these are things you're likely to have.

The first point might seem non-obvious: after all, aren't LLMs able to churn out more or less as much stuff as you would ever need? Not quite: while you might be generating slop content to get eyes on you for whatever reason, that slop still has to pass a few minimal bars. It has to be at least readable enough that non-discerning people will read through it, it has to have a compelling enough initial hook to get people to start reading it in the first place, and it has to be writing on a subject that's actually useful for you to have your name attached to. Even with an LLM, that still takes some work: you have to think of a minimal prompt about something that you think that people will read, and where reading the article will make people more likely to behave in the way you want them to. You then have to at least skim through the article to make sure it passes basic sense and relevance checks, prepare it for publishing and then publish and promote it. And you have to produce this in sufficient volume that slop of this kind makes up at least a significant fraction of the total writing on the subject you're competing for, which means publishing a lot: four or five artifacts of differing sizes a day might be what you're looking at to get the results you need. Even with an LLM, that's a full-time job for several people, especially given that if you miss any of these steps you end up with stuff that's simply incoherent and doesn't even meet the goals of flooding the zone with slop, because you'll wind up publishing stuff that's irrelevant to your audience and will thus get ignored.

Even if you can generate minimally adequate content in enough volume to make this work, if you aren't big enough in other ways your slop will struggle to be seen. Most people using the slop strategy have assets that they can rely on to get their slop seen: they have existing domains with enough authority that even if their slop doesn't show up as the first search result, some of it will show up in the first two pages. They're large enough that they can use black-hat or grey-hat SEO practices (read: almost all SEO) to build their domain authority without being flagged, they have large extant social media accounts that they can use to spread their slop around and they have email lists large enough to get at least some eyes on the stuff (and thus some backlinks) without too much difficulty. In short, they can use a commanding dominance in other resources to forcibly create a similar level of dominance in the content space by using slop.

As a small company or an individual, there's very little of this playbook that you can actually put into action. The amount of slop that you have to put out to reach anything close to saturation is far more than you could produce even with access to a whole fleet of LLMs (this isn't simply a question of me thinking that this is the case: I've experimented with getting Claude to generate social media infographics and such from my extensive collection of blog posts, and it just doesn't work: the volume required is way too high and it usually doesn't even meet the minimum quality bar for putting the slop playbook into action). Even if you do somehow generate enough stuff (by lowering standards, throwing too much time at the problem or both), you likely don't have the social media following, email list or domain authority to make use of the slop, and building that honestly requires you to produce a lot of stuff thatisn't slop before trying to compete using the slop playbook could work, even a little. Without that, your options are almost all bad. You can replicate some parts of the slop playbook through the use of bought email lists, social media follows and aggressive SEO, but if you do that, you're arguably doing something both less effective and morally worse than the slop playbook: you're engaging in grifting or scam-like behaviour. A lot of people do this, but in general, unless you're selling gas station dick pills, this isn't going to work very well.

On the whole, then, there's really no place for volumetric slop, whether LLM-generated or organic, in your playbook. If you're small enough and are writing or making other media to get business/get a job/raise your profile in your industry, your only option is to compete on quality.

Gain insight first, then put it in words

As a small consultancy or a single-person one, you compete on quality by producing a relatively small amount of work that is good enough that it can't be ignored. The goal isn't so much to use a scattergun approach to get a little bit of attention from a wide range of people, some of whom will be foolish enough to give you what you want. Rather, you're aiming to win the attention of a comparatively small niche of people who will find what you have to say properly compelling to the extent that they'll think you're a genuinely clever person. You simply can't do this with slop or banality.

Now, I might be naive and clueless in so many ways, but I've been writing (and occasionally speaking) publicly for quite some time now, and I've learned this much: the core determinant of whether people will actually, seriously pay attention to and like your stuff is the question of whether or not you have some idea or truth worth sharing that you really believe in. If you have something sufficiently insightful that your audience didn't know, or a new perspective on something they thought they knew all about, people will overlook an awful lot of clunky writing, typos, nervy speaking and presentation or any number of other things. If the point is good, useful and clever enough, people will like it, they'll share it and they'll think well of you. Conversely, if you haven't found, developed or otherwise acquired some worthy insight, no matter how good you are at writing, the whole thing will come out rather banal and you won't get attention: you'll have written something that everybody already knows and that will thus get ignored.

Obviously, insight is relative to your audience. There's no shortage of things that you, my dear readers, would find utterly fascinating that would fall completely flat on LinkedIn because people would have no idea what I'm talking about, and a lot of what does numbers on LinkedIn most likely strikes you as completely banal. There's no objective measure of whether something counts as insight, that is to say. Still, we usually have a decent intuitive idea of what the shape of insight is for our audience, and the fact that we can't pin it down objectively is no real excuse.

What this means is that, if we (small consultancies, individual consultants, engineers looking for work and so forth) wish to write for whatever reason (to get people to see us, to convince people that we're clever and know what we're talking about, even to write for money if we're that way inclined), we need to turn existing advice on its head. Existing advice for getting ahead by writing tends to be to start off by writing regularly, posting it on social media, SEO, collecting emails for an email newsletter and so on and so forth. Inasmuch as any advice is given about what you should write about, it usually focuses on writing articles about particular subjects that are of economic value to you and your clients: SEO at one level of remove, as it were. All sensible advice so far as it goes, but it won't generate insight, and if you follow it you'll probably get something fairly banal at best and outright slop at worst.

Rather, when starting with the baseline that we have to write not only something that's just good, but something with a strong enough core thread of something being said to cut through all the noise, we need to begin with insight: to begin with the idea and then make the other stuff work around it as best as you can. Most of our effort, in general, should be spent on developing ideas, trying them out, getting rid of the ones that don't stand up to testing and then executing on the ones that seem promising. Some of the writing that we produce this way will be good, some will probably be mediocre, but it'll all be making progress towards saying new, interesting and insightful things that actually materially change how people see the world (or at least the part of it you've written about).

If you don't do this, you run into two major hazards that, taken together, will probably push your writing towards banality or slop. The first of these is that without having a core idea in place before you put fingers to keyboard, you'll need to find an idea to express in your work. If you've decided that you're going to write and then start looking around for an idea, you're likely to try and choose the idea based, not on what you have to say, but on what you think other people want to hear. That never ends well, because in general us humans are very bad at knowing what other people want to hear. What you'll likely come up with is something deeply banal about AI use or a similar subject: something that simply repeats the tropes that other people have used in their own published work, lightly warmed-over, rearranged a little and published yet again. Even if people agree with the statements you've made, the lack of novelty means that they probably won't give a shit, and after skimming whatever you've written, they'll just go on to the next thing. And besides, truth and belief are compelling: it's a large part of why finding an idea that you believe to be true works so well. If you really believe in what you're writing about, it comes through in your writing, and conversely if you don't believe in what you have to say, the fakeness and dishonesty will come through and put people off: you can count on it. In short, without having an idea that you believe in already in mind before you begin writing, you will come across as dull, fake and even somewhat dishonest: none of which is useful for whatever ends you wish to turn your writing to.

The second hazard is that of script-following. Suppose that you do have a bit of an idea about what you want to write about and some belief in it: you're not in the situation described in the first hazard. However, your ideas are undeveloped and a bit vague: you have some feelings and such, but they aren't really focused down to the point of one clear, arguable idea, and yet, for all the reasons discussed above, you feel as though you want to write now rather than holding off. In that case, what often happens is that you end up repeating parts of other peoples' ideas without really thinking about them much. You might know, for example, that you want to talk about transphobia in the workplace, and you have some vague idea that it's bad, but you don't know exactly what you want to say about it and you don't have a solid line of argument. Given that you're interested in the field, you're probably going to have read a fair few articles about workplace transphobia, though, and there's the catch. It becomes very easy, when you see something clever and well-phrased, for it to sink into your head and then come out again later almost unchanged and without you thinking about it. If you read fairly widely and are observant, you can see these kinds of playbook-type snippets of text pop up repeatedly with uncannily similar phrasing across everything that's been published on the subject. Now, if you aren't careful and don't have a strong enough hold on the idea you want to communicate or what you want to say, it's pretty easy to cobble together an article-shaped object or some other writing from primarily these playbook snippets: without the idea to hold a thread together and bring discipline, you can get a franken-article with well-phrased ideas from a whole bunch of places, but no cohesion and thus no originality. These articles aren't as bad as the first class discussed, perhaps, but they're still annoying and easy to gloss over (or at least they annoy me), and importantly, whatever your goals are when it comes to writing, they're unlikely to help you meet them.

In short, there's no way of getting around the need for truth and insight if you want to write something generally compelling. One way or another, you really do have to see some part of the world in a way that's new or different and interesting, and you have to be able to form that into a coherent idea that you can communicate well. Without that, you're stuck writing with nothing to say, and in that kind of situation you'll usually just lose the contest you're fighting. How, then, might you go about writing something that can actually win the contest?

Ways to have something worth saying

So, how do we find ideas that we believe are good enough to make for good writing? The usual advice, as we've chatted about a bit earlier, isn't great, and I don't think that a lot of it really thinks that this is a problem at all. Unfortunately, finding an idea that you believe in enough that you're willing to throw some weight behind it is harder than we tend to think, and as we've established above, it is important.

Fortunately, however, all of us tend to have, by virtue of you know, regular human experience, a bunch of experiences that we can mine for good things to write about: true ideas, ideas worth believing in, sharing and defending. Here are five sources that I'm particularly fond of:

Write joined-up pieces

The general advice for people trying to write non-fiction for fun or profit is to write relatively, small, encapsulated thinkpieces, especially on places like LinkedIn. I think, personally, that this is a mistake. Individual ideas or thoughts aren't actually that interesting or useful: one idea in isolation and shorn of context is actually pretty bare and pointless, as we see a lot in software engineering. You can teach someone about, for example, the Don't Repeat Yourself principle or the Separation of Concerns principle in isolation. Individually, they can both make for great, clever-sounding blog posts or articles, and you can make a really strong case that either Don't Repeat Yourself or Separation of Concerns is the way to write software and not writing software that way will get you into real trouble. The issue arises when these two values come into conflict, which they often do (do you use one function for a bunch of stuff that requires similar logic in separate parts of the codebase and break Separation of Concerns because now two very different parts of the codebase have the same dependency, or do you have a separate function for each part of the code that can be modified and evolve differently and maintain repeated boilerplate? Is there a pattern that mitigates both concerns? If not, how do you trade them off against each other?). If you're making a habit of writing up isolated little ravioli of ideas, you lose the vast bulk of useful insight that these ideas bring, which comes from understanding them in a context of all the competing considerations that come with writing software. The ravioli syndrome of talking about individual ideas in isolation is a large part of what causes slop and banality.

If you can, then, it's often a good idea to talk in your writing, not about one idea, but about how two or more ideas interact. Where is there synergy? Where are the principles in tension? How do they interact in practice? How do the ideas sit in their context, and when are they not applicable? All these kinds of thing represent a rich vein of places where we might find insights and new ideas that often otherwise pass us by. They're often also unique to individuals much more than the individual ideas are: every article on DRY is going to read roughly the same way, while I imagine that basically every engineer would have a different take on the correct way to trade off between DRY, SOC and whatever other principles of clean code are contextually relevant.

So what ideas in your field often find themselves in tension? Which ideas are contextually interesting? What do you wish people would contextualise better?

Teach a skill to a specific audience

It's common advice for people who write for worldly gain to write lessons and to try and teach people something, and this advice has been taken to heart by many, many people: if you doubt me, all you need to do is spend five minutes looking at tech blogs on Medium. A lot of these blogs, however, are pretty banal, say much the same things as all the other blogs (that you could usually get from searching the documentation) and especially in matters technical, often give advice that isn't very good. While the advice to teach is sound, then, this advice needs to be made much more specific if one's goal is to write so that your writing will be read.

One thing that I've learned during my time in this world so far is that if you wish to teach effectively, you need to understand your students in a great deal of depth. Students differ a lot from each other, and what works for one set of people often simply won't work for another. An often-underexplored place to look when seeking ideas and insight, then, is teaching skills to a specific audience. Teaching statistics in general is altogether too vague to make for good writing: teaching statistics to professional engineers, by contrast, is specific enough to be interesting. Teaching coding skills to digital marketing people is similar: these communities have specific areas of knowledge, specific gaps in their knowledge, specific foibles and different parts of the whole field of statistics are useful to them: marketing people almost certainly want to have a strong understanding of clustering models, for example, while that might not be so important for engineers.

What group of people do you know like the back of your hand? Whom can you empathise with? And what can you teach them? Writing something teaching a skill to a specific group can make for some extremely compelling writing.

Argue for a weird point of view

In your practice of your profession, you'll almost certainly pick up an opinion or two that you strongly believe in, but that is a minority position in your work. This isn't quite the same as contrarianism for the sake of contrarianism: any two-bit content marketer can write ragebait. That isn't hard, and it's also a great way to create some properly banal takes. What I'm talking about, rather, is advocating for the minority positions that you genuinely believe in.

Take an example from my work, for example: I firmly believe that, in a well-designed application with a well-designed data model, the backend of your software should be very, very thin. In general, you should have as little between your display logic and your database as possible, and a lot of the work that we generally off-load to business logic should strictly be done in data modelling. This is a pretty radical statement, and not one that has majority backing, exactly. However, writing an article defending that position (which I should really do at some point) would immediately be highly readable. It challenges current orthodoxy in a way that makes people rethink how they design systems, and there are good reasons both to defend that logic and to oppose it (formulating complex procedural logic in a database, for example, is usually a mistake). Writing about a minority position that you hold and defending it can be extremely interesting.

So, what are some minority opinions in your field that you genuinely believe in? How would you go about defending them? What would you say?

Work across multiple fields of expertise

In the Year of our Lord 2026, it really isn't common for any of us to have spent our entire careers in the same industry and the same kind of position. Even if we're broadly doing similar things, data analysis work, for example, is done very differently from industry to industry: as I alluded above, statistics in marketing work is all about clustering models, while statistics is the field of genetics almost never touches anything of the kind but instead requires you to know lots about Bayesian models. Traffic engineering calls for a different set of skills entirely.

This is a useful opportunity when writing for gain, because it turns out that something that's taken for granted in one field or industry can be utterly revolutionary in another. For my part, I've had considerable success writing about LLMs from my perspective as a steel-and-concrete engineer: training in more physically-rooted engineering, as it happens, brings to your attention potential failure modes for the tools that simply aren't obvious from a software perspective. If you're working in marketing but have a background as a more serious writer or a journalist, you likely have a grasp of prose style that your colleagues don't: that can very much make for a meaningful article or post. There are hundreds of these things once you start looking for them.

What things are obvious in the field you've come from that are unknown in your current field? Which of those do you think your current field would benefit from being better at? Why?

Do something a little risky

Christopher Hitchens, later in his career, wrote a series of articles called "On the limits of self-improvement". Among other things, in the articles the notoriously hard-drinking, hard-smoking writer got himself a brazilian wax, booked himself into a health spa for three days, got his stained teeth dealt with at a dentist and even attempted some yoga. The articles are hilarious and I highly recommend a read of them, but they also bring into relief the last and perhaps most effective technique for finding interesting things to write about is to go out and do something, ideally a little risky and well out of your comfort zone. A writer by herself is probably a bit boring (we do, after all, choose a profession which mostly involves sitting at our desk, reading lots and then typing out a whole lot of text: not something conducive to interesting experiences): a writer who sails the Hauraki Gulf overnight in the middle of a storm is a lot more interesting.

Similarly, an AI critic is, by themselves, kinda uninteresting: most of the arguments for why LLMs are bad have been discussed to death at this point. An LLM critic who actually tries the tools for some time, makes notes and then writes more in-depth criticism on the base of that, however, becomes considerably more readable. Though I've not seen a write-up about it yet, I reckon that a heavy LLM user going cold turkey from the tools for a month and writing about that might be equally interesting, and for the same reasons. The reason, as you might have concluded from reading the previous elements, is that most interesting things and interesting ideas develop on boundaries, where things rub up against each other and find themselves in tension. It makes sense, then, that putting yourself on a boundary and creating useful tension for yourself can make for some extremely interesting writing.

You probably don't want to outright traumatise yourself when picking things to do: the ideal thing is probably something that seems a little scary and a little worrying, but that you don't think will actually kill you. If you're a non-technical person, learning how to code a small application for yourself (without the aid of LLMs) might be a good example of the kind of scale you're aiming at: if you're technical, giving a technology that you absolutely hate a serious try might be similar. If you're a front-end developer, for example, try building something in C. This kind of thing creates exactly the kinds of experience that you need to make for properly good writing.

Our job is to speak the truth

Having written the bulk of the article and come to this point, I've reached a conclusion that I don't think I had at the outset (as is common when you write things like this). Broadly speaking, what people look for when reading writing of any kind, whether fiction or non-fiction, is that you can bring to light something true that they didn't know beforehand, or that they knew but are now seeing in a different light. One way or another, you have to let people see the world in a new way.

This truth need not be objective: it can quite easily be a subjective truth, or your truth to put it in the hippyish way that really annoys me but that is nonetheless useful. But there has to be some truth, and it has to be, in some way, new. Received wisdom or facts are not, for the most part, truth except for the first time you're exposed to them. The core flaw of so much of the writing that we do for gain, and especially the kind that we see produced when we demand that people speak or write before they have any truth fixed in mind to write about, is precisely this lack of truth. Without truth, writing cannot help but come across as in some way false, dishonest or not worth our effort.

Terribly mystical, I know, but I think a good principle for any writing, the explicitly-for-gain stuff as much as the stuff that we just write for the sake of it, is not to write anything unless you have some truth to share. If you don't, what you write will come out sounding fake no matter what you try: if you do, truth will win out and you will do better at getting the gain you want by not chasing it directly than you ever would by sacrificing truth for it.