The 80/20 of Writing a Blog Post (and Why Most of It Is Not Writing)

A breakdown of where the time actually goes in writing a 1,200-word post. Plus the 4 prep steps that determine whether the writing takes 40 minutes or 4 hours.

The time auditI tracked the time on the last 12 blog posts I wrote (the ones on this site, plus 4 others for a client). The data is from my own work, not a study, and the methodology is "I started a stopwatch when I started and stopped it when I stopped," but the pattern was clear enough that I trust it.The average post was 1,180 words. The average time was 2 hours 17 minutes. Of that time, the actual writing (typing sentences into a draft) was 38 minutes. The other 1 hour 39 minutes was everything else: ideation, research, outlining, the edit pass, formatting, image selection, the publish flow, the social post. The writing was 28% of the time. Everything else was 72%.That is the 80/20. The 20% of the time that is "writing" is the visible part. The 80% of the time that is "everything else" is the part that determines whether the post ships at all.What the 80% is, exactlyFor the 12 posts I tracked, the breakdown of the non-writing time was roughly:Ideation (12 minutes avg): Coming up with the seed, deciding what the post is about, picking the angle. This is the part that, if I skip it, turns the rest of the post into a 4-hour slog.Research (22 minutes avg): Looking up facts, finding examples, re-reading my own previous posts for cross-link opportunities. For some posts this is 5 minutes. For some it is 45 minutes. The variance is huge.Outlining (8 minutes avg): Picking 3-7 sections, writing a one-sentence purpose for each, deciding the order. This is the single most leveraged 8 minutes in the whole process. The writing is downstream of the outline. A bad outline produces a bad post no matter how good the writing is.Edit pass (18 minutes avg): Reading the draft, cutting the AI tells, tightening the sentences, fixing the lede, checking the em-dash count, running the slop audit. This is the second-most leveraged step. The edit pass is where the post becomes the post.Formatting and publish (15 minutes avg): Adding the headers, the code blocks, the CTA box, the cross-links. Updating the index. Updating the RSS. Running the final link audit. This is mechanical and fast if the previous steps were done well.Distribution (24 minutes avg): Writing the social post, scheduling it, sending the newsletter mention, posting to LinkedIn if appropriate, replying to early comments. This is the part most blogs skip, and the part that determines whether the post gets read.That is the 80%. The 38 minutes of writing is the smallest single category, and it is the only category that has a minimum time below which the post is not really written. The other categories can be cut, but the writing cannot.The 4 prep steps that determine the timeAcross the 12 posts, the variance in total time was 1h 12m to 4h 50m. The variance was almost entirely in the prep, not the writing. The 4 prep steps that mattered most were:The seed has to be specific. "Prompt engineering is changing" is a vague seed, and a vague seed produces a 4-hour post. "Why my prompts kept failing at work" is a specific seed, and a specific seed produces a 1-hour post. The difference is the angle: a specific seed has an audience, a problem, and a perspective already built in. A vague seed does not. The 5 minutes of picking a specific seed saves 2 hours later.The audience has to be in the seed. "Marketing managers at small companies who use ChatGPT occasionally" is an audience. "Anyone interested in AI" is not. The audience in the seed is the audience the post is for. Without it, the post is for everyone, which is the same as being for no one. The 3 minutes of writing the audience into the seed saves an hour of "who is this for?" during the writing.The format has to be in the seed. "Three sections, each with a copy-paste example, no bullet lists" is a format. "A blog post" is not. The format in the seed is the shape the post will take. Without it, the post becomes whatever shape the writing happens to produce, which is usually a 2,000-word ramble. The 2 minutes of writing the format into the seed saves 30 minutes of "what shape is this?" during the writing.The opening has to be in the seed. "I built a blog writing factory in Python last weekend" is an opening. "In this post I will discuss" is not. The opening in the seed is the first sentence the post will use. Without it, the post begins with whatever sentence the writing happens to produce, which is usually the worst sentence in the draft. The 5 minutes of writing the opening into the seed saves 30 minutes of staring at a blank first line.Those 4 steps add up to 15 minutes of work before any writing happens. The 15 minutes is the highest-leverage part of the whole process. When I do all 4 of them, the post takes about 1 hour. When I skip 2 or more of them, the post takes 3-4 hours. The writing itself is roughly the same speed either way (about 38 minutes for a 1,200-word post). What changes is everything around the writing.What the data does not showThe 12 posts I tracked are all "good" posts. I do not have data on the posts I abandoned, the half-finished drafts, the posts I started and trashed. The 12 posts are the ones that survived. The time for the abandoned ones is harder to count, because abandonment is not a discrete event; it is a slow fade. But the estimate, from the time between "I started" and "I gave up," is 30-90 minutes of partial work, no published post. That is the real cost of skipping the prep: not just a slow post, but a post that does not get written at all.Across the 14 months of stalled writing before the factory, I would estimate I started maybe 20 posts, finished 4, and abandoned 16. The time spent on the abandoned posts is probably 25-30 hours. The output was 4 posts. The effective rate was 1 post per 6-7 hours. The factory changed the rate to 1 post per 1.5 hours. The prep steps are the largest single reason for the change.What I do nowFor every post I write, I do the 4 prep steps in a single sitting, before I open a draft. The prep is in a Notion table (one row per post, four columns for the four prep fields). The prep takes 10-15 minutes. The writing takes 35-45 minutes. The edit pass takes 15-20 minutes. The publish flow takes 10-15 minutes. The distribution takes 20-30 minutes. Total: 90-120 minutes per post, of which about 35-45 is the writing itself.The prep is not optional. I have tried skipping it. The result is always a 3-4 hour post or an abandoned post. The 15 minutes of prep is the difference between shipping and not shipping, and I have stopped pretending I can do it without the prep.What this is forThis is for anyone who writes a blog (or wants to) and feels like the writing is the bottleneck. The writing is not the bottleneck. The writing is the smallest part of the time. The bottleneck is everything around the writing, and the largest single lever in everything around the writing is the 4 prep steps. If you do the prep, the writing is fast. If you skip the prep, the writing is impossible.The honest partI have been writing blog posts for about 6 years. For the first 5 years, I thought the writing was the hard part. I would sit down to write, get stuck, blame the writing, and walk away. The 12-post time audit was the first time I had data on where the time actually went, and the data said: the writing is 28% of the time, and the other 72% is where the leverage is.If I could go back and tell past-me one thing, it would be this: do the 4 prep steps, every time, before you start writing. The 15 minutes of prep is the difference between a blog with 4 posts and a blog with 30. The writing speed does not change. The prep changes the writing speed, by changing whether the writing happens at all.Want the 4-step prep baked into a prompt pack?The Blog Writing Factory is built around the 4 prep steps. The first prompt in the pack is the prep prompt, which generates the seed, audience, format, and opening for the post. The next 9 prompts are the writing, edit, and publish flow. The prep is the highest-leverage part, and the pack is the system that does it consistently.Buy on Gumroad. $14

Send me the rough edges if you try it. I read every message.