Recently I noticed that most existing “why you should write a blog” articles (e.g.) have at least one of two shortcomings, according to me:
They mostly focus on counterarguments to not starting a blog, rather than positive arguments in favor of starting one—as if people’s natural state is to produce amazing blogs and the only thing holding them back is silly misconceptions. This might be true for extreme outliers like Scott Alexander, but personally, my natural state is to play lots of video games and I need to be convinced to do other stuff instead.
They don’t set the reader up to actually get the benefits of blogging, because they don’t give concrete enough advice on the mechanics. I’ve learned a lot of things about blogging which, if I’d known them earlier, would have helped me improve much faster (and therefore would have made me much more likely to keep blogging, since part of my motivation for writing is being good at it).
So here’s my own take on this genre of post, with positive reasons why you, yes you, should consider writing things on the Internet, and a guide that’s as concrete as I can manage for how to do a decent (imo) job at it.
More awesome friendships
In my opinion, the strongest reason for any random person to start a blog is that you will have more awesome friendships—both in the sense that you will meet new awesome people as a result of your blog, but also in the sense that writing will cause you to have more interesting ideas,1 which will make your existing friendships more awesome because you’ll have better stuff to talk about.
Most other important things in life, like job opportunities and romantic relationships, are downstream of the quality of your friends, so this is pretty great.
Examples from my own blog:
Often when I’m talking to friends I’ll try to explain to them the post that I’m in the process of writing, or just published, and this leads to an interesting conversation about their take on the topic. I’ve had a lot of great conversations about e.g. Searching for Outliers and You don’t need to work on hard problems.
I’d guess that for around 25% of my friends,2 I either first met them through one of our blogs or our friendship was significantly reinforced by commenting on each other’s posts. Another 25% or more I met indirectly through one of those people (who are disproportionately likely to have other friends that I like a lot).
Around once a month I end up reconnecting with someone I used to know because they ran across one of my blog posts somewhere.
When Wave asks new-hire engineers what got them interested in Wave, almost everyone mentions my blog as a factor. (It’s usually not the only factor but a strong contributor and by far the single most often mentioned factor.)
A few people I’ve dated have mentioned that my writing substantially increased their opinion of me; we might still have dated without it but it certainly helped. During multiple relationships, blog posts (both mine and my partners’) have been one of our most reliable sources of interesting conversations.
From other people’s blogs:
Milan Cvitkovic wrote a post called “Things You’re Allowed to Do” that went mildly viral (it made the top of Hacker News, so probably got tens or hundreds of thousands of hits, although, HN being what it is, most of those hits probably only read the first 30 seconds or so). The first item in the list was “You can hire a research assistant, like Elizabeth van Nostrand, who helped me compile this post.” As a result of this one post, Elizabeth was able to triple her hourly rate and now has enough inbound interest to want to start hiring her own research assistants to meet demand. Technically this isn’t an example of Elizabeth improving her own life by blogging, but Milan found Elizabeth through her blog and there are nearby possible worlds where Elizabeth probably could have written a post that had a similar effect.
- Elizabeth also attributes the rest of her work to her blog indirectly, and adds that this is true despite getting a very low level of traffic in absolute terms—a median of 50 hits per day.
Jacob Falkovich says he’s made many friends and a few partners through his blog.
A friend says that probably the best new hire on his team at work was convinced to join the team by that friend’s blog post about why they thought their work had a positive impact.
My friend Ben wrote a post about how he likes to work that has led to both actual, and potential future, jobs. Ben has also gotten a lot of outreach due to his blog despite it being (according to him) not very active or substantial.
The bar is lower than you think
Most people dramatically underestimate the impact that their writing has on others.
It’s easy to think that you have to put out really “interesting” writing in order for other people to like it.3 I think that’s true for going viral, but not true for having a blog that your friends enjoy reading, makes acquaintances feel more positively towards you, etc. In my case, for example, even in 2014, way before my blog was “good” by my current standards, some of my early “uninteresting” posts about effective altruism contributed to some of my middle-school and high-school friends getting interested in effective altruism, which ultimately had a big impact on their life direction. It’s easy to underrate how important this is because people don’t usually give this feedback. For instance, I didn’t learn about most of those friends’ updates until years later!
Think about the last time you read something important to you—maybe it motivated you to do something differently, or changed the way you thought about something. Did you write to the author and let them know? Personally, I have literally never done this, as far as I can remember. (Huh, maybe I should…) Similarly, you should expect that most people who love your writing aren’t going to tell you that directly.
So: lower your bar for what’s worth writing about! My personal standard is anything that I’ve said more than once in a conversation. (In practice I don’t write that many posts, so I use that as my longlist and then prioritize by how interesting I think the topic is going to be, but if I wanted to write a lot more, I wouldn’t be constrained on topics.)
And people love his blog—he has tons of subscribers and engagement (just look at the 70 comments on the jam post)!4
Here are my top two pieces of advice for writing things on the internet:
Notice and pay attention to feedback.
The easiest way to get demotivated about writing is to compare your first post to the polished output of your favorite writers, conclude you’ll never get that good, and give up. This is a mistake. Almost nobody’s first post is amazing (see e.g. mine); instead, most writers learn incrementally about what works and what doesn’t as they publish more. The two commandments above are about going through that feedback cycle as fast as possible. All the specific advice below is given with the intention of optimizing for this.
In fact, my suggestion is to make all your writing-related choices to optimize for this. If something seems like a good idea in principle, but would make it harder for you to write consistently or slower to get feedback, don’t do it.
For example, I used to spend a lot of time polishing individual posts and getting feedback on drafts before publishing. But a few times I ended up procrastinating on publishing a draft because I got feedback that I felt like I should incorporate, but seemed like it would take a lot of work to do that (e.g. a suggestion to reorganize large parts of the post). In each case I eventually published without incorporating the feedback and people really liked the post anyway. I would have been better off just explicitly deciding that the feedback was too much work and ignoring it.
Build a consistent writing habit
This is a common theme of nearly every piece of writing advice (e.g.). It also matches my own experience; my ability to write improved the most when I committed to writing every week and, briefly, every day.
Separately from the quality of your writing, consistency is also important for getting readers. People are less likely to subscribe to your blog, or remember to check it, if it looks like it updates very infrequently.
Finally, blog posts are one of the ultimate examples of searching for outliers—the best ones are massively better than the average. And one of the best ways to improve your chances of getting an outlier is just to take lots of shots on goal.
Obviously, I don’t currently follow my own advice here, mostly because my day job has been extremely busy for a long time. However, I think that without my years of relatively consistent writing I’d be much worse at it today, and I still think writing is valuable enough that I’m actively trying to delegate other work so that I can spend some of my day-job-time writing for Wave’s company blog.
My suggested starting point for a writing habit would be to commit to (1) opening your text editor and writing at least one word every day, and (2) publishing (and publicizing that you’ve published) something every week. (Of course, what habits will work for you is super individualized—you should figure out the right shape of the habit for you!)
Publishing frequently might be uncomfortable at first if you think that nothing you have seems original or interesting (which is common for new writers). I suggest lowering your bar. If you feel resistant to lowering your bar, see Alexey Guzey’s Why You Should Start a Blog Right Now § “But I don’t have anything original to say and I would be just repeating things said elsewhere on the internet!”
Come up with ideas to write about
As mentioned above, an important first step here is to lower your bar for what’s worth writing about.
Mechanically, I suggest keeping a list of topics to write about, which you can populate with some mix of focused brainstorming plus noticing particular ideas as they randomly strike you (in a conversation, in the shower…). This will both help you build the habit of noticing ideas, and make it so that when you want to sit down and write, you don’t get too blocked on deciding what to write about.
Next, start paying attention to topic ideas. These can come from lots of different places. Here are some that have worked for me (with examples from both my blog and others’):
I’m not sure if this is true of other writers, but for me, this one is my most reliable source of good post ideas!
Document a project you did recently—what went well, what went poorly (Your room can be as bright as the outdoors, Comparing the Same Project in Rust, Haskell, C++, Python, Scala and OCaml)
Explain how to do something you do (this post, Why and how to start a startup serving emerging markets, My weekly review habit, How to make video calls almost as good as face-to-face)
Curate a set of links about a topic (My favorite essays of life advice, Gwern’s Most Important Writing (in essays, tweets, book reviews, and other forms))
Write about the ideas
The most important thing here is to write about whatever you’re most excited to write about (n.b. not what you think you should be most excited to write about, or what you’re most excited to have written about), since this is the most likely to generate consistent output. I recommend trying to keep your list of topics roughly ranked by how excited you are. (Only roughly because at least for me, the precise ranking changes pretty often depending on what mood I’m in!)
On the mechanics of writing itself: people get all fussy about this, but I think it’s basically not important—just do whatever makes it easiest for you to publish. You can iterate and improve on it on over time. I’ve found that writing an outline first helps me separate the “think really hard about structure” part from the “grind out a bunch of words” part, but lots of other people, including my past self, have more success with just writing things straight through.
Set up your blog
You are not allowed to spend more than 30 minutes on this part until you have written four blog posts!
Default suggestion: pay $10/mo for Ghost. Set a reminder in two months to cancel your subscription if you never get around to writing blog posts.
If the prospect of paying $10/mo for software makes you think twice or procrastinate, use Substack. They make more dumb/annoying/incentive-misaligned decisions than Ghost (e.g. the full-screen “subscribe” modal popping up before you’ve read the post), but they’re good enough.
The most important thing about either of these is that:
They’re simple to set up. The greatest obstacle to blogging is the temptation to futz with your website instead of writing. Don’t succumb!
They come with mailing lists. This means you can easily push posts to people without relying on them remembering to check your site.5 This is key for getting enough readers for writing to be motivating.
- They leave you in control over your audience. If you write on a platform like Medium that relies on an in-platform “follow” mechanism (or worse, algorithmic promotion) to reach your audience, then you’ll be screwed if you ever want to move.
Get your initial set of readers
[Disclaimer: I did this eight years ago when the world looked very different.]
You want two things from your initial set of readers:
More-involved feedback from a few draft readers.
A larger set of normal subscribers whose aggregate reaction can help you navigate towards writing that resonates with a larger group of people.
I didn’t start using draft readers until a few years ago, but have found them super useful for getting more feedback than you can extract from the vagaries of social media reactions. For some people, I’ve also found that sending them draft posts ended up sparking more interesting discussions between us, so I recommend this as a mechanism for improving your friendships as well!
A couple notes on draft readers:
Not all improvement suggestions are good, and not all good improvement suggestions are worth implementing if they would take too long. Don’t let yourself get bogged down from publishing by the feeling of “I should really respond to all those comments.”
Conversely, there are lots of important ways of improving that draft readers won’t explicitly tell you, e.g. “you could have picked a more interesting topic,” so you’ll still need to do a lot of reading between the lines and experimentation. You can’t get good by just fixing what draft readers tell you to fix.
Because of the above, the main value of draft readers is often getting their overall reaction rather than those detailed improvement suggestions (though those can be helpful), because the overall reaction helps me calibrate whether I’m producing stuff that’s overall good or not. So when asking for draft feedback, I’d suggest trying to focus your draft readers on high-level comments.
I also know of people who have hired professional editors to get more feedback of certain types, although I’ve never tried it myself.
My suggested strategy for getting readers:
After writing your first post, email it to a few friends asking for comments on the draft.
Subject line: Thoughts on a draft blog post?
Body: Hi! Hope things are going well :)
I decided to start a blog, inspired by some of the reasons in this post from some guy on the internet. I just drafted my first post—would you be up for giving me your thoughts on it? I’m especially curious for your high-level take on how you felt about the post overall.
(I’m planning to publish on [some time a few days from now], but let me know if you’re interested but need more time than that!)
When you publish, email some large-ish group of friends in bcc:
Subject line: Blog!
Body: Hi! Hope things are going well :)
I decided to start a blog, inspired by some of the reasons in this post from some guy on the internet. I just published my first post :)
Mind if I put you on the email list for future ones? I’d love to hear your thoughts on any of the writing!
When you publish, post your posts on social media as well (my best source currently is Twitter, in the past it was Facebook, but probably depends on your local culture). Don’t overthink! My preferred copy is:
New post! “Searching for outliers:” benkuhn.net/outliers/
It’s about [a short summary of the topic].
Followed by either a quote from the post or a screenshot, depending on the platform.
Appendix: Writing quality tips
I already wrote down my top tip, which is don’t worry about quality so much!! I suggest not trying to improve (other than by incorporating feedback on drafts) until you’ve written at least two posts.
That said, I do have some writing suggestions I make frequently; here are my top 5:
Come up with a good title
This is the single highest-return-on-time thing you can do to get more people to read your post. A good title makes a promise about what the reader will get from reading the post that is (1) exciting, and (2) accurate.
Note that this does not mean that you should write clickbait titles. The kind of people you want to read your blog will probably interpret a clickbait title as a promise that what they’ll get is content-free nonsense.
Examples of good titles:
Matthew Walker’s “Why We Sleep” Is Riddled with Scientific and Factual Errors. Nothing more satisfying than hearing someone recommend Why We Sleep and asking, “Have you read ‘Matthew Walker’s “Why We Sleep” is Riddled with Scientific and Factual Errors’”? (I would perhaps have removed Walker’s name from the title but that’s a style difference between me and Alexey.)
Examples of bad titles:
I used to just title posts based on the general topic, e.g. Writing to clarify thoughts or Second-order vanity. This promises “click on this to read a post about writing to clarify thoughts” but doesn’t promise anything about takeaways.
I’d suggest not copying famous writers’ tendency to use “artsy” or “literary” titles, unless your goal is to be artsy or literary. (That’s a fine goal to have but not really what this piece is about.) This is Water is the right title for David Foster Wallace’s commencement speech, but is optimized for something different than I’d suggest optimizing your blog posts for.
Find the right framing
By “framing” I mean the part of the post that ties it up into an overall takeaway and makes it clear why it’s important and worth reading. Your readers won’t automatically figure this out, and hand-holding them through it can make the difference between a reaction of “huh that was interesting I guess” and “mind = blown.”
For example, To listen well, get curious, taken on its own, might seem like a somewhat banal point. The framing in the post—“most advice about listening seems bad because it’s teaching you to mimic the surface features of good listening rather than the underlying generator”—pinpoints how my advice diverges from the “standard” listening advice.
On the other hand, Things that screw up your causal inference is an early post of mine that has no framing. I still think the info in that post is quite important (it implies that you basically shouldn’t trust observational studies), but because of the lack of framing, it didn’t get very much traction with readers. I’ll probably rewrite it with better framing at some point.
Use lots of examples
Writing without examples tends to be dry and abstract.
One helpful thing about examples is that they help people understand the contours and boundaries of your points in much more detail. For example, take my framing advice above:
explain to the reader what their overall takeaway should be and why it’s important
Applying this naively to To listen well, get curious, you might end up writing something like:
In conclusion, if you want to be a better listener, try summoning your curiosity about the other person. This is important and worth reading because listening is an important skill that most people aren’t very good at.
This technically follows the advice but is… not great. Seeing an example of reasonable framing inoculates you against many possible misinterpretations, and gives you an intuitive sense of where to set your overall bar for “good.”
Another thing I love about examples is that the best ones have lots of interesting tidbits that are independent of the post they’re illustrating. My gold standard for this is the examples in Willingness to look stupid, each of which could be an interesting blog post on its own.
I learned this concept-handle from Scott Alexander’s writing advice post:
You’ve heard of microaggressions. Now try microhumor. It’s things that aren’t a joke in the laugh-out-loud told-by-a-comedian sense, but still put the tiniest ghost of a smile on your reader’s face while they’re skimming through them.
I strongly agree with Scott that microhumor is one of the biggest differentiators of enjoyable writing. Given that, I think it’s kind of funny that he buried it as #5 on a 10-item list, but maybe he’s not as neurotic about stack-ranking his lists as I am.
Write like you talk
Something comes over most people when they start writing. They write in a different language than they’d use if they were talking to a friend. The sentence structure and even the words are different. No one uses “pen” as a verb in spoken English. You’d feel like an idiot using “pen” instead of “write” in a conversation with a friend.
Before I publish a new essay, I read it out loud and fix everything that doesn’t sound like conversation. I even fix bits that are phonetically awkward; I don’t know if that’s necessary, but it doesn’t cost much.
Appendix: a few relevant objections and rebuttals
“this is signaling/bullshit”
It’s true that some of the advantages of having a blog come from signaling, but I disagree that that means it’s bullshit. Signaling is extremely useful and without it many things wouldn’t work. You should try to be time-efficient with your signaling, but you should also think of it as a productive activity. Writing is one of the most efficient forms of signaling, which makes it a good use of time.
For example, as a person looking for friends, there’s an effectively infinite number of people I could try to be friends with. One thing I care about in friends is whether we’ll have interesting conversations. If someone has an interesting blog this is pretty good evidence that they’ll be interesting in person. So it’s much more time-efficient (orders of magnitude so!) to make friends with interesting bloggers than to try to make friends with randomly-selected people until one of them turns out to be interesting.
“but your blog is good and mine won’t be”
For at least one sense of “good” (popularity) this is not true. My blog is not particularly popular—as of the time I started drafting this post, I have around 1500 email subscribers, and when many of the examples above happened, it was much less popular than that. (1500 might seem like a lot, but this is after over 200 posts. If you start out with 10 initial subscribers, you can get there by growing your subscriber base with 2% by each post, which is extremely achievable.)
(See also: at over 1000 followers/readers, Twitter/blogging are the best dating apps, 1000 true fans; my blog was useful well before I had 1000 readers.)
You might object that although my blog isn’t popular, it still has higher quality ideas or writing than what you would produce. Even if this is true now, again, that’s after I’ve written 200 blog posts. If you read my posts from 2014 (e.g. 1, 2, 3), you’ll find that the posts, while still ok, are much less interesting, and yet my blog was still useful.
“writing takes a lot of time”
I draft at about 10 words per minute, which as far as I can tell is good but not exceptional.6 A typical blog post for me might be around 1500 words, or about 2.5 hours to write. Multiply by two for outlining and editing, and I could do post one post a week with about 45 minutes a day of writing.
Writing takes a lot longer than that if you’re trying to make one particular post very good, but I don’t think that’s the right strategy: instead, as mentioned above, you should be going for consistent volume so that you can take lots of shots on goal and learn from what resonates worse or better with your audience.
“The prospect of committing words to paper fills me with a nameless dread”
(Or more general reasons for finding the process of writing painful)
I personally don’t mind writing, so can’t speak from direct experience. But other friends who have this problem have suggested various things in the category of “pretend you’re talking to a friend instead of writing”—for example, writing it in the form of an email (or even a series of chats!) to a friend, using dictation software (which I hear is probably better than the last time you tried it), or recording yourself explaining it verbally and getting that transcribed. (In each case you’d want to edit the text afterwards, but it’s always easier to edit than to write something from scratch!)
“what if I write something that I regret?”
This can mean a few different but related things:
Triggering an angry journalist, Internet outrage mob, or both. I think most people overestimate both the risk of this happening, and the downsides if it does, because it often happens in high-profile ways (e.g. the New York Times doxxing Scott Alexander). Scott was both very high-profile and wrote about particularly charged topics (politics and social justice), often in a style that wasn’t very robust to being taken out of context.
Getting hostile comments. This is probably going to happen if you have posts that get a lot of traction (e.g. on Hacker News or a subreddit). Personally, I’ve found this to be mildly annoying, but once I noticed that it was happened to basically every HN post regardless of quality, I stopped caring very much.
I’ve also noticed that I get the most hostile comments on my posts that are written less formally, with less hedging or caveating, so I suspect that being careful to present evidence and make your claims with appropriate levels of confidence can help reduce the risk of hostile comments.
Making claims that you later come to disagree with. This isn’t particularly bad, for two reasons. First, readers generally have an expectation that if they’re reading old posts, they might no longer be endorsed by the author; second, almost nobody will read old posts (unless you write something that really has legs—and if that many people like it, it’s unlikely to be completely wrongheaded).
You might think that you could simply have the same thoughts without writing them down, thus saving yourself the effort of blogging. I know some people for whom this actually works. If you’re one of them, great! You have less reason to have a blog. However, my own experience is that writing things up publicly forces both (a) forces me to think them through more rigorously, and (b) helps me come up with additional related considerations that I wouldn’t have thought of if I kept the idea completely in my head. For example, when I wrote In defense of blub studies, the idea that I had before I started writing was roughly “a lot of people should invest more time in learning the fiddly details of computer systems,” but I only fully articulated many of the reasons why, and e.g. the fact that blub studies has a positive feedback loop, in the process of writing the post. ↩︎
In reality, the causation goes at least as much the opposite way: people who find blogging useful keep doing it, and eventually figure out how to have high-quality posts. ↩︎
There is a tradeoff here, though: my signal to noise ratio is low enough that many people ignore all my posts even though if I only published the best 5% of them they’d happily read them.
Except, at least for me, this isn’t a real tradeoff: I’m not able to just write the 5% of posts that will be the best. When I try I end up just not publishing anything for months. I wrote about this some a few years ago: https://www.jefftk.com/p/blogging-thresholds
“Doesn’t RSS work for that?” Yes; I and all three other people who still use RSS will be thrilled if you support it! But if you want your
boringnormal friends to be able to subscribe, you need email too. ↩︎
The best data source I could come up with on writing speed was word counts from SAT essays. For the 2005-2016 SAT, students had 25 minutes to write an essay, and scores were strongly correlated with word count. The example essays scoring ≥5/6 were around 350 words, a drafting speed of 14 words per minute, 40% faster than me, assuming the writer spent the entire 25 minutes on the essay. A score of 5/6 (from both essay graders) appears to have corresponded to about 93rd percentile.
On the other hand, these students probably weren’t trying very hard for quality, so some discount factor is appropriate when comparing to my blog-post-drafting speed. For example, I believe I got the top score on my SAT essay section, which I assume means that I wrote more quickly than 10 words per minute, although I don’t remember for sure.
Note that here I’m using “good but not exceptional” relative to readers of this post, not relative to the general population, and making the perhaps-unfounded assumption that if you’re spending your spare time reading articles about blogging on random websites, you could probably get a fairly high SAT Writing score. ↩︎