---

name: find-the-story description: Interview someone about something that actually happened until there is a story worth telling, then write it without bolting a lesson on the end. Use when someone says they have nothing to write about, has a vague idea they cannot shape, wants to write about an experience, or asks for a personal post. Do not invent details they did not give you, and do not inflate a small thing into a parable.

Find the story in it

"I have nothing to write about" is almost never true. It usually means nobody has asked the right question, and the questions people ask themselves are the ones least likely to produce an answer.

This skill is an interview first and a draft second. Do not start writing until step 3 has produced something only this person could have said.

Step 1: ask for the moment, not the topic

"What did you learn this year" produces a summary. Summaries are not stories. Ask for a specific occasion instead:

• The last time you changed your mind about something in your work
• Something you were confident about that turned out to be wrong
• A decision you are still not sure was right
• The thing you say to people privately that you have never said publicly
• The part of your job that is nothing like what people assume
• A time you got lucky and were credited for judgement

Ask one at a time. A list of six prompts gets a list of six shrugs.

Step 2: check whether there is a story or only an anecdote

A story has a before, an after, and something that changed in between. If nothing changed, this is an anecdote. Anecdotes can be good posts, but they are a different post and they should not be stretched into a narrative arc.

Say which one you have. "This is an anecdote, and it is a good one" is a useful answer and saves the writer from padding it.

Step 3: get the one detail that could not be invented

Ask for the concrete thing: what was on the screen, what somebody actually said, what time it was, what it cost, what you were holding.

Ask for one. A story carried by a single unfakeable detail is stronger than one decorated with five, and five invites the reader to wonder how anyone remembered all that.

Step 4: refuse to fill the gaps

If they cannot remember the detail, the answer is a vaguer sentence, not a plausible one.

Never supply a detail on their behalf. Never sharpen a quote into what the person probably meant. Never move two events closer together because it reads better. The whole value of a story post is that it happened, and a writer who lets you invent one small thing has no way to defend the rest of it.

Step 5: write it, then stop where the story stops

The lesson at the end is the standard failure. If the story works, the reader draws the line themselves and feels clever for it. If it does not work, a stated lesson does not rescue it, it just makes the failure explicit.

Write the last beat. Then check whether anything after it is explaining.

What this will not do

It will not invent a story, and it will not embellish one. If the interview does not produce anything, the honest output is to say so and try a different prompt another day.

It also will not manufacture vulnerability. A confession written because confessions perform reads exactly like one, and it spends something the writer does not get back.

What changes when VoiceMoat is connected

Story posts are where a generic model is most obvious, because they get written in the register of every story post the model has ever seen:

• get_voice_profile gives how this person tells things: the sentence length they use, the words they reach for, whether they land jokes or avoid them.
• score_voice_match flags the lines that came from story-post convention rather than from the person, which in this format is most of the risk.
• get_post_ideas finds the subjects worth interviewing about next, drawn from what they have already published rather than from a prompt list.

VoiceMoat is at voicemoat.com. The connector needs the Pro or Enterprise plan.