The messy process behind producing a single post on 'The Calculated Risk'

1:23 AM. Thirty-seven browser tabs open. On the left monitor, three nudge theory paper PDFs are stacked on top of each other. On the right monitor, a blank Google Docs page with nothing but a blinking cursor. Next to me, a cup of coffee gone cold. This is the typical scene when one of my blog posts begins.

Sometimes people read a post on The Calculated Risk and say, "That's pretty well organized." I appreciate it, but if they saw the actual process, they'd probably take it back. Today, I'm going to crack that process open.

It Starts by Falling Down a Rabbit Hole

Topics usually don't come from planning. They start with an offhand remark from a professor during class, a comment on Reddit, or a question like "Is that actually true?" that comes up over dinner with a friend. The problem is I can't stop there.

The gig economy post I published last month was like that. It started when a friend who works as a delivery rider in Vancouver said, "Honestly, I don't even know how much I make in a month." I went home after hearing that one sentence and started looking for articles on the income structure of platform workers, and before I knew it, I was deep into how "present bias" in behavioural economics affects gig workers' financial planning. I checked the clock: 3 AM. My first class the next morning was at 9.

At this stage, I don't write anything. I just read. I open tabs, skim paper abstracts, play YouTube lectures at 1.5x speed, and sometimes I'm simultaneously queuing up for ranked Valorant. I'll read a paragraph of a paper during the loading screen and come back to it after the round ends. This inefficient multitasking is a kind of thought-organizing process for me. It feels like my brain is processing information in the background.

Something Too Messy to Call an Outline

Once I've hit around fifty tabs and get the feeling that "I can probably write something with this," I open Google Docs. Then I dump out sentence fragments as they come to mind. This is not an outline. There's no structure, the grammar is a mess, and Korean and English are mixed together. A note like "mention anchoring bias in this section" sits right next to "Vancouver housing prices are insane."

What takes the longest at this stage is choosing the angle. The same topic becomes an entirely different piece depending on where you look at it from. Say you're covering AI and employment. There's the "AI is stealing jobs" perspective, the "AI is creating new jobs" perspective, and the "that whole debate oversimplifies the problem" perspective. I usually lean toward the third one. It's just my tendency to think both sides are a little right and a little wrong. The downside is that it makes the piece longer. Capturing nuance requires space.

This is where the battle with myself begins. The first draft is almost always too cynical. When I was writing the stock market trend analysis piece, the opening sentence of my draft was "Most retail investors have no idea what they're doing." It might be factually true, but if I publish that as is, I just come off as arrogant. The tone of my blog is "cynical but fair," and that sentence isn't cynical. It's rude. Finding that line takes a lot of time.

Negotiating with Self-Censorship

After writing a draft, I usually let it sit for about a day. If I reread it right away, I'll gloss over parts thinking "this seems fine," but a day later, those same parts start to look like "this isn't quite right."

While it's sitting, I do something completely different. I go to a café near Robson Street and people-watch, browse indie film festival programs, or read city council documents on Vancouver urban planning. It might look like procrastination, but it surprisingly helps the writing sometimes. I'll be reading urban planning documents and the concept of "incentive design" connects to the core framework of the marketing piece I was working on. When moments like that happen, I quickly jot them down in my phone's notes app.

When I open the draft again a day later, the thing I do most is delete. About 30 percent of the first draft gets cut. What mainly gets removed falls into two categories: parts I included to sound smart, and tangents unrelated to the flow of the piece. Even if I picked up an interesting fact from a behavioural economics rabbit hole, if it doesn't connect to the core argument, I cut it. Reluctantly.

Then comes the truly painful stage: checking whether my argument is fair. I have pretty strong opinions, so I sometimes make the mistake of downplaying positions I disagree with. That's why I seriously ask myself, "What would someone on the other side say if they read this?" If I spot a vulnerability that can be rebutted, I include that rebuttal directly in the piece. It makes the writing longer, but I value the reader feeling that I'm "fair" more than feeling that they "agree" with me.

The Final Gate

In the final editing stage, I read the piece out loud. During hours when my roommate isn't home. Reading aloud immediately reveals where the rhythm of a sentence feels off. Sentences that look smooth on the page can leave you out of breath or unsure where to pause when you say them out loud. When that happens, the sentence is too long.

I decide on the title almost last. Since the blog name The Calculated Risk itself carries a touch of irony, I try to match the title to that tone. I don't use exaggerated clickbait titles. If I use titles like "This Is All You Need to Know" or "The Shocking Truth," I attract readers who ignore the nuance of the piece, and those readers won't read to the end anyway.

Right before I hit the publish button, I always hesitate one more time. That moment of "Should I really post this?" There has never been a single post I was 100 percent confident about. It always feels like I could have researched more, found a better way to phrase something, or that there's still a gap in the logic somewhere. But I've learned over dozens of posts that a perfect piece never gets published. If it's reasonably solid and reasonably honest, that's the time to publish.

Honestly, I'm writing this post at 2 AM right now, and when I reread it tomorrow morning, I'll probably want to rewrite half of it. But that's how this blog works. Behind every clean result, there's always a messy process, and I've become less and less ashamed of that mess.