---
type: "Love post"
title: "The Idea - Write - Edit process: Escape the never-ending writing loop"
description: "A three-stage process with stopping points that frees the ideas stuck in your storage and gets them out onto your website."
resource: "https://cuongthach.com/y-viet-sua/"
tags:
  - "blog"
  - "en"
  - "Publishing process"
status: stable
generated:
  by: "process:cuongthach-okf-export"
  at: "2026-10-20T00:00:00Z"
---
# The Idea - Write - Edit process: Escape the never-ending writing loop

**Idea - Write - Edit** is a process that fits into three disciplines: **write the script and commit to it, pour it all out without judging, and edit only to cut extra words**.

When these three jobs get mixed together, you write, edit, then redo the ideas, and you never finish a post. Split them into three steps with stopping points, and you break out of the loop of editing again and again. The ideas that have sat in your storage too long get out onto your website, told through your own experience.

## Why do you keep writing but never publish?

![Why do you keep writing but never publish?](https://cuongthach.com/_astro/illus-vong-lap-viet.en.xC1BB_7K_ZTKTiY.svg "Do it all at once and posts go in circles")

**Because you are working on the idea, the writing and the editing all at the same time.** With no process, you go by feel: you write when you feel like it, then fix the ideas once it is written, then find it weak and write it again. The post keeps going in circles and never makes it out.

### A writing loop with no stopping point

I call it the **writing loop**. It goes like this:

- You make an outline but **do not lock it**.
- You write, and write all you want, and spend a lot of effort.
- You look back and the post feels too scattered. Of course it does: it is scattered because the ideas were never locked.
- You edit all you want. The post is too long and still scattered, so you edit more. Now it is too short and does not get the ideas across.
- You end up editing the ideas too, without noticing. Suddenly the post and the title no longer match: *"Wait, where did my message go?"* You go back to the outline.

Just three rounds like that and a post takes **three times as long** to publish. A hard post you edit up to ten times, which means you are **ten times slower than others**.

And that is before we get to how it feels. Write and rewrite forever and you lose heart. With nothing to hold on to, you start to fear writing, and **fear breeds procrastination**.

You have ten posts waiting to go out, but one post takes four days. Not because you write badly, but because you tore it down and started over ten times. By the second post, you are worn out. One slow post sets off a **domino effect** across your whole content plan.

### Two traps that stop you from publishing

**The first trap: perfectionism.** I lived in that loop. I suffered from something called **perfectionism**: I had an idea, wrote it down, found it not good, then tore it down and started over, many times.

My posts might score a 9 while other people's scored only 7.5. But I published **half as fast** as they did, and that is not good. What I lacked was not words, but **a process**. From those stumbles, I boiled it all down into the Idea - Write - Edit formula, so I would have a frame to follow.

**The second trap: sitting and thinking instead of getting to work.** You take a topic and start to think: *"What do I write? What should the ideas be? Why do I need this idea? Am I missing anything?"* You think and think and think, and in the end you do nothing at all.

Thoughts that stay in your head never get written down; one interruption and you forget them. You forget, think again, then forget again, and four or five days later nothing has come out. If it comes out at all, you got lucky; usually the post stays stuck there forever.

### Ideas that sit in storage forever

The end result is that your ideas **never get out**. You collect a lot and you have many good ideas, but they all still sit in your storage: Notion, Google Docs, Obsidian, all kinds of documents.

If you do not publish, nobody knows. And nobody knowing is very dangerous. **An idea that sits in storage, no matter how good, is a failure.**

## What is the Idea - Write - Edit process?

![What is the Idea - Write - Edit process?](https://cuongthach.com/_astro/illus-no-ra-nen-lai.en.DNhgUE6K_ZTKTiY.svg "Write: pour it all out. Edit: only cut")

**Idea - Write - Edit is a three-step process for publishing a piece of content.** Once you have the [Know - Trust - Love content funnel](https://cuongthach.com/biet-tin-yeu/) and a full plan, Idea - Write - Edit is how you carry out that plan. Every post in the funnel, whether a Know post, a Trust post or a Love post, goes through the same process.

Picture each post as a movie. At each step, you wear a different hat, and the next step only starts once you hold the result of the step before:

| Step | The hat | In your hands when done |
| ---- | ------- | ----------------------- |
| **Idea** | Screenwriter, the person who writes the script for the content | A locked outline |
| **Write** | Actor, who takes the script and acts it out right away | A rough draft, at its longest, not yet edited |
| **Edit** | Editor, who cuts it the way the director wants | A finished post, ready to publish |

Each type of post has its own kind of outline, because the readers of each type want something different. But **the only difference is how you make the outline**; writing and editing are the same. This is a process for publishing, not a mold that forces every post to look like every other, so it will not make your posts dull or cookie-cutter.

> [!NOTE] Not a formula for making sales
>
> Idea - Write - Edit helps you **actually publish** a post. Whether that post sells is the funnel's job.

### What is the Idea step in the process?

**The Idea step is when you create the outline, which is the script of your post.** It is the step that takes the most time, and also the step most worth your care.

**The outline is the script of your post.** The point of a script is **consistency**, not saying whatever you feel like. When readers look at the outline, they grasp three things right away:

- The message of the post.
- The overall idea of the whole post.
- Whether the ideas flow from one to the next.

People get it at a glance, and AI also understands your post very quickly.

Just like a movie, if the script is wrong from the start, stuffed with too many messages, or its small ideas are scattered and do not flow, then no acting, however good, can save it. So the end goal of the Idea step is **a truly clean, locked outline**, and you try not to change it again.

**Being different starts with the idea.** Your post has to be different in its ideas first, before we even talk about stories or words. Then within each idea, you are different once more: you have **two layers of difference**. Being different in your ideas makes your post **one of a kind**. There are two cases:

- **An idea nobody has talked about:** you are the first to talk about it, so of course the post is one of a kind. Your job is to become the first person to talk about that idea.
- **An idea others have talked about too:** you make it one of a kind with your words, with your stories, by bringing your own self into it.

This is also where **people beat AI**. The Idea step is the step AI does worst, where it can least compete with people. A good script is not a blend of ready-made content patterns; it calls for something original and unique, and it needs a person to create it first. Treat each post as **a creative work**, and know that the script is the only thing that makes your work different, not the words.

A post with good ideas gets praised right away, often before people finish reading. Today a reader's instinct is to filter four things: **the title, the description, the opening and the main content** (which is your outline). One skim of those four and they know right away this is a good post.

### What is the Write step in the process?

**The Write step is when you lay out your outline as fully as you can.** You work from the outline you locked in the Idea step, expand each big idea and each small idea, and write or say everything you need to say.

**Pour it all out; it does not need to be good yet.** At this step, you do not care about fixing or polishing the words. First, you have to make it clear to yourself. Of course you still aim for your customers to understand, but first you have to get out **everything in your head** about each idea.

This is a stage of intense focus, **100% on letting your ideas pour out**. You do not try to:

- Fix the words.
- Fix the stumbles or the repeats.
- Switch to your customers' language.
- Persuade anyone.
- Stuff in keywords.

Repeating an idea two or three times is fine. Leaving things out, stumbling and starting again is no problem either. Pour it all out so you know **where you are still missing something and where you stumble**.

**Turn off the internet when you write.** You cannot write and search at the same time. When you write, you turn off the internet, open no documents and do not go back to read your notes: it is just you and the page.

This is when you dig deep into everything you understand and know: your own processes, memories and experiences. Every time you go online to check something or learn more, your train of thought gets cut off, and this is exactly when it needs to pour out in full. Checking and research must be done **before you write**, in the Idea step.

### What is the Edit step in the process?

**The Edit step is when you trim your draft into a finished post.** You edit what you just wrote, so your customers understand it and like it.

**Editing is cutting, not adding.** Your draft says a lot, but it is not yet in your customers' language. In the Edit step, you do three things:

- **Switch to your customers' language.** Just because you understand a term does not mean your customers do. Say it plainly, in their words, use words that are already widely known, and do not rework them in ways that are hard to follow. The rule is **never make your customers think**.
- **Rearrange the words.** Where you repeated yourself or put things badly, you fix it, pull it together, and turn long sentences into short ones.
- **Compress it.** Cut the repeats, the confusing parts, the parts you understand but your customers do not, and the extras not tied to the idea. While you pour out the draft, you just write it, because it seemed a little bit related; but when you reread it by instinct and it feels extra, cut it, no second thoughts.

You only cut and trim; you **never stretch the post out**.

> [!TIP] The rule for a long post
>
> **Add ideas, not words.** A post with 10 ideas is long on its own. A post with only 3 ideas that you pad out, holding back from trimming just to reach a word count, is the wrong way.

Usually a 10,000-word draft comes down to about 4,000 to 5,000 words after editing.

Editing is also when you format: draw tables and add pictures, so your customers can read with ease. But remember, **to compress, you need enough raw material**. The draft has to be thick enough to squeeze something good out of; squeeze a tiny one and the post ends up paper-thin.

**Edit the content, not the outline.** In the Edit step, you do not change the titles or the small ideas; those were already polished in the Idea step. Changing titles over and over slows you down, puts you behind, and builds a habit of fiddling with everything. And when you rewrite a title, you do not always get the meaning right: **change just a few words and the title means something else**.

When you write, you write to the original idea. When you edit and rework the wording of that idea, it can turn into a different idea without you meaning it to; the odds of this are very high. You think it has not changed, but in the end it has, and the post starts to **fall out of tune** between the title and the content below it. Changing a title, to put it plainly, means changing everything under it too. So: **once the outline is locked, that is it, no more changes.**

## How to carry out the Idea - Write - Edit process

```mermaid
flowchart TB
  accTitle: Each step has a stopping point, and if writing stalls, go back to research
  A[Locked outline] --> B{Does it pour out?}
  B -- Not yet --> C[Research more]
  C --> B
  B -- Yes --> D[Rough draft]
  D --> E[Compress the opening first]
  E --> F[Finished post]:::focus
```

**Each step has one main job and one stopping point you must reach.** Only once you reach the stopping point of one step are you allowed to move on to the next.

### Step 1 (Idea): Lock the outline

**You start with very little: a title, or an idea that fits in one line.** Do not sit and think. Take it apart right away with three warm-up questions:

1. *What is it?*
2. *How does it work?*
3. *Why does it matter*, to you and to your customers?

Answer these three from your own experience and what you already know, and do not go online yet, not at all. These three questions are the test: if you cannot answer them, you do not understand the topic deeply yet, and that post should not be written yet. Only share what you truly understand and have run into yourself.

As you answer, your thoughts will not stop. Like a car that is already rolling, just let it keep going: dig deeper into each idea that pops up, pour it onto paper or record it, until you have twenty or thirty, even forty or fifty rough ideas spread out in front of you.

Then you **filter with your core message**. A core message has three parts: who your customer is, what their problem is, and what your solution is (see also the guide [how to create a core message for a post](https://cuongthach.com/docs/know-trust-love/guides/create-a-core-message/)).

Keep only the ideas that directly serve that message. Putting everything in waters the post down and loses its consistency. Usually forty or fifty rough ideas get boiled down to about **six to ten main ideas**.

Last, you sharpen the titles of the big ideas: short, easy to understand, direct, no riddles, no slang or metaphors.

The detailed moves for this step are in the guide [build an outline from a one-line topic](https://cuongthach.com/docs/idea-write-edit/guides/outline-from-one-line/).

> [!IMPORTANT] The stopping point of the Idea step
>
> You reread the outline and find the script good enough, and it flows well enough.

### Step 2 (Write): Pour it all out in one go

**Between the Idea step and the Write step, there is a handover.** You look back at the outline you just locked and check yourself: some ideas you have 70% down, some you have forgotten 30% of, some are still hazy. Wherever something is missing, you research it right now: look up material, watch videos, read books to fill your blind spots completely.

But fill them with your own understanding. Not by copying and pasting into a note so you can paste it into the post later. You have to understand it for the words to pour out naturally.

Once your blind spots are taken care of, start writing while the "fire" is still there. This time you do not scatter your thoughts; you hold tight to the big ideas locked in your outline and dig really deep. Turn off the internet, no looking things up, no AI. Your only job now is to pour everything you understand about each idea onto the page, without worrying about wording or spelling.

> [!IMPORTANT] The stopping point of the Write step
>
> You pour out, in one go, through every idea in your outline. If you stall on any idea, it means you did not prepare well enough at the handover: stop, research until you have enough, and only then keep writing.

### Step 3 (Edit): Trim and compress the post

**In the Edit step, you edit in order of priority: compress the opening first, then compress the main content and finish the ending.**

The opening is the hardest part and takes the most effort. In the Idea step, you cannot know yet how you will open. Only when all the content has poured out in full in the Write step can you see the whole picture and boil it down into a few strong opening lines.

The opening must be very short and go straight to the point. Readers today only scan 4 things: **the title, the description, the opening and the table of contents**. If the opening rambles or the table of contents is scattered, they leave the page before they get to read what you wrote below.

For the main content, you only need to reformat it so it is easy on the eyes (headings, bullet points, tables) and compress the wording so it is crisp and sticks close to the big ideas. Remember: the draft has to be thick enough to squeeze something good out of; force a squeeze on too little material and the post ends up paper-thin. And most important: **never rework the outline** at this step.

The detailed moves for this step are in the guide [edit a draft and compress the opening](https://cuongthach.com/docs/idea-write-edit/guides/edit-a-draft/).

> [!IMPORTANT] The stopping point of the Edit step
>
> You hold a finished post in your hands: you understand it, your customers understand it, and the words are lean, just enough, and straight to the point.

## Why must you go in the order Idea, Write, Edit?

![Why must you go in the order Idea, Write, Edit?](https://cuongthach.com/_astro/illus-leo-nui-chang-nghi.en.DkaT-VQl_ZTKTiY.svg "Like climbing a mountain with rest stops")

**Because the idea matters most, the idea has to come first.** Many people flip it around: write first, edit next, and only then make an outline. That is exactly the door into the loop.

### With a good script, any performance is good

You have a topic and jump right into writing. Only after writing and editing do you ask: *"Hmm, does this serve my core message?"* Then you redo the outline; you find ideas missing, so you write more; once you add them, it feels a bit off, so you change the title; once you change the title, you change the ideas above it again. An **eternal loop**.

At last, worn out, you say: *"Fine, version 1 is good enough, just publish it."* And the post is bad from version 1. Version 2 will probably repeat the same thing. Make ten versions and, if you are lucky, one of them is good.

But when you lock the ideas and the script is already good, **however you perform, it is good**. In a movie with a good script, the actors no longer matter; even if you present it a little badly, people still like it.

On the other hand, if you write all over the place, act all over the place, long-winded and scattered, nobody can save the movie. If you present something and people do not understand it, they rate your content low.

Going in the right order, I usually **do it only once**. There is no version 2; version 2 is only an update of whatever has changed. You are not making a marketing page to A/B test, then C/D test. Writing content needs speed and accuracy, it needs to be steady and sure: once you publish, you can rest easy.

### A process ends procrastination

With no process, **you only write when you are happy, and when you are down you cannot write**. You write for two hours straight on a burst of inspiration, then a phone call, a problem, a task cuts in. When you come back, you have lost your direction, lost your fire, and you cannot pick up the thread.

Your energy drops and your drive slowly fades. Repeat that enough times and at some point you quit: *"Forget it, I am going out."* When you write by feel, what you feel is drained, like a battery running down.

With a process, you no longer care whether you are happy or sad. You finish the outline first. Once you have the outline, you just follow it and write, 100% focused, forgetting everything else, and you reach a state of **flow** while writing.

The internet is off, so nothing pulls you away. At the Edit step, you edit what you already wrote. Your feelings no longer control you. **A process cures procrastination in publishing**: publishing starts to feel ordinary, you just do all three steps and you are done.

It feels like **climbing a mountain with rest stops**. You climb to one peak, rest, then climb the next. You know where you are, where the next rest stop is, and that you are heading the right way.

A process also tells you **which step trips you up**. With no process, all you can do is complain: *"I publish so slowly, just one post a week, and I do not know where I am stuck."*

With a process, you see it right away: *"Oh my gosh, I have spent four or five days on the outline."* So you trip at the Idea step, while writing and editing are still fast.

Once you can pinpoint the problem, you can fix it very fast, and once you want to fix it, I believe you will. Knowing where you are stuck, you **improve extremely fast**.

## Where do you use AI in the Idea - Write - Edit process?

![Where do you use AI in the Idea - Write - Edit process?](https://cuongthach.com/_astro/illus-ai-tang-bang.en.Bbj0r6V2_ZTKTiY.svg "AI serves, and you always decide")

**All three steps of Idea - Write - Edit depend almost 100% on people.** AI is a tool that goes along with you and serves you, but everything it does is based on your own data.

### What AI does in each step

You do use AI in Idea - Write - Edit, but in the right places, and you are always the one who decides:

| Step | AI does | You do |
| ---- | ------- | ------ |
| **Idea** | Groups and pulls apart the ideas you have spoken; compresses big ideas written at length into easy sentences; fine-tunes the titles of the big ideas, the small ideas and the post | Pour out all the ideas, then sort, shape and cut them down to the final outline |
| **Write** | Turns your recording into text, removes filler words, splits it into paragraphs and formats it on its own | Say or write it all out, without looking anything up |
| **Edit** | Spots repeated ideas, suggests what to compress or compresses it for you; suggests highlights for the opening | Check and approve, trim the final version yourself, write the opening yourself |

At the Edit step, AI reads very fast, so it edits very fast too. It does the rough trim, and you trim the final version; or if it does not trim as well as you, you trim it yourself. It depends on your strategy and your skills. But AI **does not change the titles of your big ideas**, because the outline is locked.

The parts that decide the outcome only people can do. At the Idea step, your instinct creates an outline that is truly human and truly gripping, because **only people know what people like**. At the Edit step, the opening needs empathy; only people can feel for each other, and without empathy you cannot write the opening.

So if you want to be faster, **train yourself**, do not go buy a better tool. Train enough, and what used to take you 2 or 3 hours now takes just 1. There is no such thing as buying a tool, plugging this process into it, and sitting idle while AI does the work.

### You never win a race of tools

In the age of AI, publishing is actually even harder. You write badly, and AI writes better than you. It came along later than you, yet it is better, and it keeps getting better. With no process, no formula and not enough practice, you will lose to it, then give up and hand it everything. In the end you **do not get any better**, because you never trained at anything, so how could you?

Hand the writing to AI and it reads like it is good, like it is fine. One click and it is out, no hour or two spent, and all you do is sit and tweak. But it is all show, with nothing of you in it. Content that does not come from a person lacks a human touch and real experience, and it brings nothing that lasts, especially when it comes to sales.

You cannot win the tool race either. If you have AI, other people have AI too. Some of them also have tricks for getting the most out of the tools that you do not have, and you lose, not because you know less, but **on a trick**.

And racing on tools costs money:

- Some people spend 100 or 200 dollars a month.
- Some people use 4 tools for 1,000 dollars.
- Some people use 10 or 20 tools.

Race that way and you will never win, your whole life. But **racing with your brains, racing with yourself** costs nothing. With a process and practice, even someone using 100 tools is only on par with you.

### Readers and Google can smell AI

**Readers are getting more and more careful.** One whiff of AI, one trace of AI, and they judge your brand on the spot: *"This guy is no good. If he cannot even write, how can he sell?"* They believe a brand that leans on AI cannot be trusted.

Overseas there is even a wave of boycotts; people call it **AI slop**. Try putting yourself in the customer's shoes: you read a messy post and you can tell AI wrote it. Would you keep reading? If customers do not read it, why write it at all?

Algorithms also favor people and feelings. Google keeps getting tougher on [content mass-produced just to pull up rankings](https://developers.google.com/search/docs/essentials/spam-policies#scaled-content), and sooner or later that kind of content gets rated low.

> [!WARNING] The risk has three levels
>
> - Getting dropped from the index.
> - Losing trust for your whole site and everything tied to it.
> - Losing trust for each post.

In the long run, people who know how to write will outrank you, just because you had the experience but did not write it down, and gave it to AI to write instead.

Look at the big AI companies. I have not seen a single AI company use AI to write its posts. Read OpenAI's posts and you will see. They do not hold their meetings with AI either.

They are AI companies that do not use AI to write their posts, because if an AI company used AI to write, people would not read it, and the ones who did would slam it. So why would an ordinary person use AI to write posts? **You are being trapped**, because the people selling tools want your money.

### Reading well matters more than writing beautifully

**Making content does not require a gift for writing.** Trying hard to write beautifully can sometimes backfire: too many figures of speech, metaphors or slang will slow down how fast both readers and search algorithms take things in. Customers today are very practical: they need answers that are fast, clear and straight to the point.

**80% of a post's success lies in the Idea step**, and this step is guided by **your reading instinct**. When you look at an outline pulled tight around your core message, your reading skill lets you feel right away: does this line of reasoning flow yet, is this script gripping. Whatever flow of content convinces your instinct will also convince your customers.

**This skill does not need a talent you were born with.** If you have the habit of reading books, reading the news or watching good content every day, that instinct has already formed in you on its own. I am not a beautiful writer myself; I just focus on making things really clear.

**The highest goal of a post is to help customers understand**, not to make them marvel at the words. A compliment on your writing is not the same as an order. People decide to buy when they:

- Clearly understand the solution.
- Trust your expertise.
- See real value for themselves.

## Practicing Idea - Write - Edit in real life

![Practicing Idea - Write - Edit in real life](https://cuongthach.com/_astro/illus-ren-luyen-khung-gio.en.DFtjuAl-_ZTKTiY.svg "The only way to speed up is to repeat")

**There is no way to speed up the process except doing it again and again.** You repeat it until you no longer need to remember what the Idea step does, what the Write step does, what the Edit step does; you just follow that default flow.

### Publish steadily on a tight time budget

Most solo business owners do not have long, unbroken blocks of time to sit and write straight through from morning to afternoon. Your time is usually torn into scraps: sometimes 30 free minutes, sometimes 1 hour, sometimes 2 hours. Without a process, those short stretches slip by for nothing.

Say **you can only scrape together a few free hours each week**. The Idea - Write - Edit process still helps you get posts out, thanks to its **separate stopping points**. You finish one step for good, wrap it up, then come back with peace of mind and carry on in your next free slot:

| Free slot | What you need to do |
| --- | --- |
| **30 minutes** | Dig into the 3 questions and finish the outline (Idea step) |
| **30 minutes to 1 hour** | Write in one go, or record yourself pouring out the rough draft (Write step) |
| **2 hours** | Trim, compress and format the finished post (Edit step) |

When time is very tight, you can record yourself instead of typing in the Write step: your voice pours out faster, then let a tool turn it into rough text before you move on to editing.

### Keep practicing, and get feedback from people walking with you

Do not expect it to stop feeling heavy right after you finish this post. At first it still feels heavy, because you lack content and are not used to the process yet. But after one, two, three, four weeks of sticking with it, your content slowly fills up, and you get used to the work and know the process by heart before you notice; everything gets lighter and becomes a reflex, with nothing left to memorize.

You will go much faster **with people walking alongside you**. When you get it wrong, someone fixes it right away, and only when you dare to share will someone dare to correct you. Bring the posts you write with Idea - Write - Edit to [the HXN Tribe](https://hxn.vn/bolac/) for real feedback, instead of fumbling around on your own for two or three months before you get the hang of it.

The skill you build here is **a publishing skill**, not a writing skill, and it has been proven on me: an introvert who stumbled in exactly these places, then boiled it down into a formula. It is not hard; it is only hard while you do not understand it. Understand the three steps clearly and practice enough, and it will fit you on its own.

## Keep going with me in Write to Sell

The Know - Trust - Love funnel gives you a content strategy, and **Idea - Write - Edit** is how you make that strategy real. If you do not publish, every plan stays on paper forever.

Master this process and you will not only publish steadily, you will also build the strength to master any content format on your own: something no AI tool can do for you.

If you want to go further, to package your product and build a sales machine from your own words, I will see you in [Write to Sell](https://cuongthach.com/write-to-sell/).