Idea · Exercise · 17 Aug 2026 ·
The False-Start Log: Fifteen Minutes That Salvages a Dead Idea
废稿日志：十五分钟救活一个死掉的动机 — 用中文读这一篇
Every abandoned idea is a small data set. Most writers throw it away and start again from zero, which means they re-learn the same lesson a dozen times. This drill takes fifteen minutes and turns three dead ideas into a reusable list of what actually stops you.
Song structure
- VERSE
- CHORUS
- 1
- 5
- 9
- 13
Step one: collect three corpses
Open your voice memos, your notes app, or the folder of half-finished files. Pick three things you stopped working on in the last six months. Do not pick the ones you are proud of — pick the ones you genuinely abandoned. Fifteen minutes total, five per idea.
- 1. Two minutes: write down exactly where you stopped. Not 'the chorus', but 'second line of the second verse, the one about the phone'.
- 2. Two minutes: write what you were trying to avoid. Almost always it is one of four things — a chord you cannot play, a subject you do not want to say out loud, a note you cannot hit, or a decision you did not want to make.
- 3. One minute: write the smallest possible next action. Not 'finish it' — 'play the verse in a different key' or 'cut the phone line'.
Step two: name the pattern
Across three corpses you will usually find the same two avoidances repeating. That repetition is the entire point of the exercise. It is not a personality flaw; it is a habit you can design around.
| Repeated avoidance | What it usually hides | One counter-move to try |
|---|---|---|
| Stopping right before the second verse | You do not have enough material; the idea was a hook, not a song | Write the second verse first next time |
| Rewriting the same four bars | The melody is fine but the harmony is boring | Change one chord in the loop and stop editing the melody |
| Never recording a demo | Fear of the vocal being heard | Record the demo with the melody hummed, no words |
| Starting a new file instead of continuing | The current one needs a decision you are avoiding | Write the decision down and give yourself two options only |
Step three: build the salvage pile
For each of the three, spend the last minute doing the smallest next action you wrote. Not finishing — just the one action. Sometimes that is enough to reopen the file. Often it is not, and the idea stays dead, but the action itself is now proof that the blockage was smaller than it felt.
Keep a single note called the salvage pile. Every dead idea goes there with its one-line next action. When you sit down with no ideas, you open that note instead of a blank page. A pile of thirty half-songs with written next actions is a far better creative bank than thirty blank new files.
Why this works better than forcing the idea
Forcing a dead idea usually means rewriting the same section until it goes flat. Auditing it separates the idea from the avoidance, and avoidance is the part that repeats. Once you can name your two standard avoidances, you can pre-empt them at the start of the next song — for example, by writing the second verse before the first, or by choosing a key you can actually sing on the first take.
Carry these
- Write down where you stopped, what you were avoiding, and the smallest next action — five minutes per dead idea.
- Three corpses will reveal two repeating avoidances; those are the things worth designing around.
- Keep a salvage pile note; it beats a blank page.
The exercise
Take three abandoned ideas. Five minutes each. Write where you stopped, what you were avoiding, and the smallest next action. Then do the action for sixty seconds. Keep the results in one file called the salvage pile.
Questions that come up here
How many dead ideas should I audit at once?
Three. Two is not enough to show a pattern — you will read it as a one-off. More than four and you are doing admin work rather than writing. If you have hundreds of fragments, sample the three you abandoned most recently, since those are the avoidances you are still carrying.
What if the pattern I find is 'I get bored'?
Boredom is usually a symptom, not the cause. Look one level down: bored at the same point every time means a structural gap, bored after the first recording means the song lacks a second event, bored when practising means you can already play it. Write the more specific version next to the vague one.
Same workbench
- IdeaBreakdownTwo-Chord Seed: A Song Built on Em-C, and Where It Needed a Third
两和弦种子：一首建立在 Em-C 上的歌，以及它在哪里需要第三个和弦
E minor (Em–C), capo 4 to put the melody between D4 and B4 · ABCB
- IdeaTechniqueVoice Memo Triage: Turning Four Hundred Fragments Into a Short List
语音备忘分诊：把四百个碎片变成一张短清单
n/a — this is a cataloguing method · n/a
- IdeaExerciseWalk and Hum: A Forty-Minute Filter With No Phone
边走边哼：四十分钟不带手机的筛选
No instrument — find the key at home afterward · none enforced
- IdeaTechniqueMelody Before Words: Writing a Tune on a Neutral Syllable
先旋律后歌词：在中性音节上写一条旋律
F major, sung on 'da' throughout the first pass · decided after the melody, not before
Spotted something wrong, or know a record this example comes from? Write to [email protected], or use the contact page.
The two languages here are written separately rather than translated line by line, so the examples and the working vocabulary differ where the language does.
This is a working notebook on songwriting craft, not professional advice. What follows describes what happened in particular songs — test it at your own desk rather than reading it as a rule for your music.