I have started, by honest count, eleven novels. I have finished one. The math on that is bleak until you realize the real failure wasn't the ten I abandoned — it was that I spent two to three months on each before discovering the idea couldn't hold a reader's attention, including mine. I was test-driving cars by buying them. There had to be a cheaper test, and it turned out the cheapest test is a cold open.
Why the opening scene tells you everything
An opening scene is a brutal little diagnostic. If you can't find one moment in your premise that's interesting enough to put on the first page, your premise is a Wikipedia article, not a story. A good premise generates openings; a thin one generates throat-clearing. So instead of outlining for a week, I now spend forty minutes generating openings and reading them cold.
The tool I use is Turn a One-Sentence Logline Into a Cold-Open Scene, and the move that makes this work is the scene moment field. I take one logline and run it five times, changing only that field: the inciting discovery, the first lie, the moment of no return, the quiet domestic scene that hides the bomb, the aftermath. Five entirely different first pages from one idea, in about ten minutes, all from Claude Opus 4.8, which is consistent enough across runs that I'm comparing the idea and not the model's mood.
A Tuesday this actually happened
Last month I had a logline I was in love with: a marriage counselor realizes one of her clients is her husband's mistress. Delicious, right? I thought it was my next book. So I ran five cold opens. Four of them were exactly what I expected — tense office, the recognition, the held breath. Competent and dead on arrival. But the fifth, set on 'the morning after she figures it out,' opened with the counselor making her husband's coffee exactly the way he likes it, two sugars, and the prose found this awful tenderness in the routine, and THAT was the book. Not the confrontation. The performance of normalcy. I'd have never found it if I'd outlined, because outlines plan the obvious scenes.
Here's the uncomfortable flip side: I also ran my treasured 'AI gains sentience in a smart fridge' comedy idea through the same five-scene test and every single opening was a chore to read, including the ones I tried to like. The idea was a tweet wearing a novel's coat. Killing it cost me forty minutes instead of two months. That is the entire value proposition.
The reason this works where outlining fails is that an outline lets you cheat. You write 'Chapter 1: she discovers the affair' and your brain fills in a competent, abstract version that feels promising precisely because it isn't real yet. An actual opening scene can't hide. It has to commit to a specific morning, a specific room, a specific first line, and committing is where thin ideas collapse. The cold open is the moment the abstraction has to become prose, and prose doesn't lie about whether an idea has blood in it.
The five scene moments I always test
- The discovery — the inciting incident played straight. Your baseline.
- The lie — someone hides the truth from someone they love. Tests whether your stakes are interpersonal or just plot.
- The aftermath — start AFTER the obvious dramatic beat. This one finds books the others miss.
- The quiet bomb — a calm domestic scene the reader knows is about to detonate. Tests tension.
- The point of no return — the choice that can't be unmade. Tests whether your character has agency or just gets things done to them.
What I do with the survivors
When an idea survives the five-opening test — when at least two of the scenes make me want to keep reading — I know it has enough surface area to be a book. Then I take the strongest opening and feed it into my Build a Chapter Beat Sheet From a Premise prompt to map the first chapter, and I'm off. The cold opens stop being disposable and become the actual seed. Nothing is wasted: the audition becomes the foundation.
One honest caveat, because I don't want to oversell a forty-minute trick: the test is necessary but not sufficient. A great opening can still front-load an idea that has no second act. What the five-opening test reliably catches is the opposite failure — the idea that's all concept and no scene, the premise that sounds good in a logline and dies the instant it has to become a specific Tuesday. That failure mode is the one that's cost me the most months, so catching it cheaply has changed how much I actually finish.
I'm down from eleven abandoned novels to zero abandoned novels this year, not because I got more disciplined but because I stopped committing to ideas that couldn't pass a forty-minute audition. Grab the Turn a One-Sentence Logline Into a Cold-Open Scene prompt on Prompt Dock, take your most precious unwritten idea, and run it five times before you let yourself fall in love. It's the kindest cruelty you can do to your own writing.