It takes me 2 hours.
I am not joking. Usually, I don't have it "planned" before I started to write. Sometimes I do when it comes to technical related topics, but anything usually comes down to me jotting it down on dev.to and refining it from there.
When it comes to writing anything but technical, I tend to have an idea of what I want to write during the weekend, but nothing written down. I just hope I remember it in a couple days. If not, oh well I figured something out lol.
For curiosity, in this case, I decide to write this piece since @ben release some news on AI Disclosure for dev.to. It gave me a thought: "How long yall have taken to write one dev.to article?"
For Technical topics, I do write things down during the week. For example, on my article on "Git Gud", I write a list of git commands I tend to use more often on a sticky note. Then, I documented it on dev.to and go from there. This is more of a straightforward example, but overall, it takes me a week to have a technical piece finished. Compare to anything but technical piece, it's 2 hours lol.
Sometimes, I refined my article by me re-reading it (just like right now) or have my friend peer review it. My friend enjoys what I have written on dev.to and probably will join, even though he is not a tech person.
I don't use AI on my writing (as mention many times before), mainly because I want to improve on my writing and not falling into the habit of relying on AI tools. Trying to get out of the habit as a developer since I have notice I used AI more than I should for learning sake.
How long does it take for you?
I am curious since I notice some people publish everyday or once a while. I am wondering about your process of writing an article on dev.to and how long it takes you? Would love to hear your experience and your thoughts about the use of AI to help you on the process!
As a side note, this took me 40 mins to write this whole thing out, including finding images for this post lol.
Top comments (51)
It really depends on the article (duh), but let me break down:
If the article is very deep technically or in any other way, it can take as much as 3 months to put together.
But I've written articles in a few hours, the ones that don't require too much research but still send an important message.
That's fair. Anything roughly in a few hours make sense, but 3 months is crazy. Have you written an article that took months to do??
I have, this one for example took months because back then vibe coding was still an uncharted territory and before writing anything I wanted to get all the information and try it myself.
Writing it up took just a few days, but the research, observation, and practice parts took literal months.
I don't write articles unless I'm absolutely sure I know exactly what I'm talking about, that's the main reason why I can be really slow sometimes.
And then I have this one which I wrote up in just 2-3 hours, because I had already made notes for myself. All I had to do was turn those notes into an actual article.
Makes sense research and other notes takes more time than writing itself.
Depends very much how long it takes me to write an article. Some started spontaneously in 1 hour, but I usually still refine those later, or delete them. More crafted ones can take or or two days, counting ideation, editing, translation, and later updates, both here on DEV.to and on Open-Mind-Culture.org.
I still don't use AI much, except for helping me to structure and decide about intereseting angles and relevant content niches to avoid tumbleweed posts or reiterating what other already wrote a hundred times before.
Thanks for sharing Ingo.
Can you give me an example that shows this in detail? How do you determine if something has been already written? Do you have to hunt articles and read them through?
I don’t really have a fixed timeframe.
Sometimes I can bang out a hot take in a couple of hours. But even when an article is meant to be provocative, I try not to just scratch the surface — there’s usually quite a bit of thought behind it.
If I had to put a number on it, though, I’d say most of my articles take somewhere between 4 and 20 hours from start to finish. Sometimes longer, very rarely less.
That's interesting. It makes sense to get it out under couple hours. I tend to have a good idea from the get-go but there are cases where it takes me a few days if I am busy during that time where I am not able to do it in one sitting. Thanks Pascal :)
Most of them a few hours, spread across a couple of days. That's with AI help though. When I wrote them by hand, I would spend all week writing one. So AI has shaved off a couple of evenings 😁
Fascinating. How would you write by hand? Do you plan before hand or just ball?
I'd have a general idea, but just go with whatever came out. Then I'd end up moving things around and trying to organize it in a way that makes sense to humans that don't live inside my head. That part is what usually takes longest. 🤣
The answer is...it depends. 2 hours sounds about right.
Most of the time i start on google docs and dump my thoughts, so i basically have a backlog of many docs with word vomit on different topics.
If i feel a topic is attention worthy or good material for explaining, i will probably create the sections i want to cover and make the word vomit a bit more structured. Some times the long format doesnt suit my content and this is when i turn to figma to make my tidbits (which it takes surprisingly WAY MORE time)
What it takes me the most time in long format (even after living abroad for 7 years) is to correct my grammar in English. Greek is a way more descriptive language and things doesn't sound "correct' translated 1-to-1. Also find sentences that convey the meaning that i have in my mind is generally hard. For this i use mostly grammarly, i have tried to use ai tools in writing but i dont like their style of wording.
Yay for my hero academia cover, i am in third season
Hey Alexandra! Thanks for sharing! Your structure is better than mine the way I read it. Also, yea, English is a language for most of us. For me, it taken years for me to understand English such as past, present, future tense, understanding sarcasm, etc. English was not my first language, even though I grew up in the US. I started learning via TV and shows and go from there. Works out surprisingly and still do to this day but less than before.
I feel like a fraud, but I never watch my hero academia actually. I saw clips over the years on YT and just learn characters from WIKI. I do want to watch MHA and I know they release the final season, but I am hesitant...because...you know...it's MHA
Regardless, thanks again Alexandra :D
I grew up in a country that the English level is... bad, we learn English in school and get certificates, but it is very common to not learn proper pronunciation, i also learnt great deal with movies and series. I didn't speak English often until my mid 20's. I was so embarrassed to ask anything in my first trip in Europe xD
There are wayyyy more engaging anime to watch, MHA is fun but a bit repetitive. I like feel good animes so i suggest to everyone these days to watch "The witch hat atelier" :)))
Just had to look it up and looks interesting. Have to put that in my long list of Anime to watch, though how is MHA repetitive since I didn't really expect it to be repetitive. I would expect other countries' English level is bad since it's not usually the first language. Can't blame really.
I can't spoil you the plot for you, but for me it got meh in the third season. Maybe because I like a specific kind of genre.
That's a difficult question! 🙀
For me, one article per week is probably the fastest pace I can realistically maintain, and each article usually takes at least a few hours.
For example, my current DSA View View series takes quite a bit of time because I try to explain the algorithms carefully and also create GIFs for the visualizations. I think each article takes more than two hours in total.
On the other hand, articles like my comparison of OpenAPI generators are completely different. That involved technical research, actually trying the tools, comparing them, and verifying the results, so from start to finish it took me around a month.
I have a lot of things I'd like to write about, so I think I'll just keep going at a slow pace.
Even with AI assisting me, writing a technical article properly still takes quite a lot of time. 😸
That's great to hear Nyaomaru! I am wondering if there is a piece you did that took more than a month to do? Regardless, thanks for sharing :)
I don't think I've ever spent more than a month just writing an article! 😸
Of course, building OSS projects or tools themselves can take much longer, but for the writing part alone, about a month is probably the longest so far. 😼
Honestly, with trying to get everything right and all the polishing it can be a whole day (I don't rush it) if it's not technical. I am not writing the article continuosly for 24 hours but...
Sometimes it would take my whole morning to reasearch and put together.
And if it's very technical it could be from 5 days to a week, maybe even two if I have to run the code again to get screenshots.
Hey Daniel! Hope you are well. Obviously, I would HOPE you are not writing 24 hours lol. What's consider "very technical" for your case and what article took you that long? Regardless, thanks again Daniel!
So it was the one where I wrote about how I built the 2D Racer in Python. I had to structure it, hunt for screenshots, look at data formats that will help people to understand(I think I added a table), and ways to not make it borring. So All this while also knocking out my tickets during the day. Then I draft it and then when I next have time, sometimes the same day sometimes the next day or two... I do a spell and grammar check before finally posting.
Usually 2-3 hours or 1-5 days.
If I get stuck on a problem where the solution was hard-to-find but straight-forward-to-solve once you figure it out, I'd usually write a quick post in a few hours to make sure that solution was easier to find for the next developer encountering that problem. I find myself writing less of these in the past year though because AI has made it much easier to find solutions for these kinds of problems.
Then there are my more technical posts. Those can take up to a week. Writing the article can take days, then editing can take days more. I'll often rewrite entire sections of the blog and the example code because I'll realize that something about the same code made a concept hard to explain. Updates to code require testing to make sure everything works. It's a long process.
Thanks Tyler for sharing! A good reasonable time length for posts.
I started to see that based on your activity on dev.to. It's a sad reality imo. I am glad you are still around hanging. Gives me more hope to see veterans like you :)
Speaking of technical, I started to see less for my end as well. I tend to write posts that are more casual than anything. Brings in a lot of human connections like this one while still talk about anything in tech.
Thanks again Tyler for sharing :)
Hello Glad to see you, I am Kane Lim from Hong Kong. I have over 10 years of development experience. I am writing this because your post was interesting.
I think the two hour estimate is actually healthy because writing is a different engineering loop from coding. I usually treat technical writing like a design review: capture raw observations first, build a logical structure, then iterate through clarity, correctness, and reader comprehension.
The most effective workflow I have found is separating ideation from composition. Keep lightweight notes during development, then turn them into an outline when the pattern becomes clear. Peer review is especially valuable because another developer exposes assumptions you no longer notice.
AI can accelerate drafting, but there is a real tradeoff when the goal is improving technical communication. If the model performs the synthesis, you lose some of the cognitive work that develops your own explanatory ability.
For me, a strong article is not measured by publishing frequency. It is measured by whether another developer can understand the problem and apply the solution.
Curious to compare writing workflows with other developers here.
You said it all. I've had very few that took up to a week to post because the writing inspiration is mostly drawn from my development process and testing. Some people are writers yet slow and other are not, yet they've got content. It took almost 2 hours to get a cover image for my last post 😂
lol. I am glad to hear that my reply was of some help.
I would like to get to know you better. Would you please contact me? t_g_@kanelim1997
Hey Kane. Will reply to your post soon but just a quick heads up, please do not insert your telegram in the comments. It is known for spam and it does get suspicious based on your activity history I am seeing from you. If anything, add to your bio. People will contact you naturally.
thx your advice
Hey Kane. As promised, here is my reply :) Also thanks for sharing :D
How would you determine if the developer can understand the problem while also applying the solution? I remember making a post about this whether the content is consider good for this case:
Debates AI use and content quality standards
How would you decide, whether the content is good or bad?
Regardless, thanks for sharing :)
It depends on how it feels. For the first 18 instalments of the ongoing 36‑Stratagems series, with the outline already set, I could finish one chapter a day at my fastest. Back then I had downtime at work and plenty of opportunities to sneak in writing. As the plot progressed, however, the pace slowed down. Right now each stratagem takes roughly 2‑5 days, it’s hard to pin down an exact timeline. Everything outside the main story can mostly be wrapped up in half a day. Most of my extra time goes into translating my original Chinese drafts into English. I cross‑check translations and localisation across multiple AI tools, trying my best to get my original ideas across perfectly in English. I only publish once several models produce comparable outputs. Even so, I still can’t guarantee every nuance comes across exactly as intended, haha. 😂This is pretty much the best I can do.
A lot of work it seems lol. Good work regardless Xuling (Saw you made a post on about "My Book 'AI, Ego & Regret'", congrats :))
Without you helping me sort out my account issues back then, this book might not have been published. Thank you for your congratulations on the new book; it means a great deal to me.👏
Some comments may only be visible to logged-in visitors. Sign in to view all comments. Some comments have been hidden by the post's author - find out more