The interviewing is most of the work. The writing is the comparatively simple act of rendering what the interviews already established.Share on X
People tend to imagine that a ghostwriter sits down with a client, has one good long conversation, and then goes off to write the book.
Some ghostwriters work that way, and I think it shortchanges the client. The books come out thin where depth was needed and general where specifics would have done the work, written by someone who understood the topic in theory and never spent enough time inside the client’s thinking. The author pays for that every time someone reads the book.
My practice is different. By the time Joe Rockey and I finished our interview phase, we’d spent the equivalent of a full workday across multiple sessions on structured conversation. The reason more time is better is what happens to a client’s thinking when someone asks them to examine it repeatedly, from different angles, with no job in the room except understanding what they’re saying.
Put a number on it. The discovery phase for a book like Joe’s runs to many hours of recorded conversation. That becomes a stack of transcripts long enough to print as a small book on its own, before a word of the manuscript exists. People are sometimes startled by that ratio. They imagined the writing was the work and the interviewing was preparation for the work. It’s the reverse. The interviewing is most of the work, and the writing is the comparatively simple act of rendering what the interviews already established. It scales with the book, though. With a client I started with recently, the whole analysis phase took one hour, because that’s all it needed. That book had nothing to do with the client personally, so I didn’t need to know anything about them. We talked about the book, and boom, we were into the writing.
After every session I make a summary document from the transcript, with AI’s help, and send it to the client. They get to see their own thinking laid out and come to the next session ready to push it further. I think every ghostwriter should do this. A client who never sees what you heard has no way to tell you what you got wrong.
The first session with any client is mostly orientation. They give me the polished version of their thinking.
What they say in presentations. What they’re proud of. The structure they’ve explained enough times that it comes out smooth and automatic. That version is the starting point. It’s never the destination. A particular kind of answer signals I’m still in orientation territory: the answer that sounds like a headline. Short, clean, a little too neat. Useful as a pitch. Not useful as the foundation of a book.
A book requires the thinking behind the headline, the cases that tested it, the moments where it didn’t work and what that revealed.
There’s a reason the polished version is so smooth, and it’s the same reason it’s so useless for a book.
The client has optimized it over hundreds of repetitions to end a conversation with a nod. Those repetitions have sanded off every rough edge and interesting exception in the name of clarity, and those edges are exactly what a book needs. A writer who stops at the pitch ends up publishing a brochure under the client’s name, and I want no part of that.
By the third or fourth session with Joe, the quality of the conversation had changed in a way I recognize from projects that produce good books.
He’d stopped presenting his expertise and started examining it, taking ideas somewhere he hadn’t taken them before. Qualifying a position mid-sentence when he heard something that didn’t quite hold. Circling back to stories he’d mentioned in passing, because they deserved more than the passing mention.
A client hired me for a policy memoir and the interviews found something else entirely. That project is in The Policy Memoir That Became a Night.
What does a ghostwriter listen for in an interview?
There’s one moment I listen for across the sessions, and it’s the moment the client surprises themselves. They follow a thought, arrive somewhere they hadn’t planned to go, and pause. Then comes some version of: huh, I hadn’t put it that way before. That pause is gold.
It means we’ve moved past the rehearsed material into something forming in real time.
What forms there is almost always more honest and more specific than anything that came before. A single-session interview never reaches that moment, because it only happens after the easy material is exhausted. It’s my favorite part of this work, and a rushed process throws it away.
Four things that came out of the later sessions
Four things came out of those later sessions that are going into the manuscript. The first is about what most managers give the people they promote:
I know how to ride a bike, why can’t you? Pedal fast and don’t fall. Well, thanks, jackass. Now what.
That’s a precise description of the leadership training most line managers receive, and I think it ought to embarrass companies. They promote their best performer, hand over a team and walk away. The second captures the entire transformation his methodology produces in two short sentences:
They don’t have to be a dick. They just think they do.
The third explains why the conventional approach to team communication fails for the majority of people on any given team:
You’re only fluent in about 25% of the personality combinations that exist.
If you can only communicate naturally with a quarter of the personality types you manage, the other three quarters are receiving a message that doesn’t land. Nobody knows why.
The manager thinks they’re being clear. The employee thinks they’re being misunderstood. Both are right. The fourth quote is about the assessment tool that makes the methodology work:
This test tells you the edges of your puzzle piece, and which pieces you should and shouldn’t connect with.
Why does the good material only come in later sessions?
None of those four lines would have appeared in a single-session interview. A person still delivering their standard presentation doesn’t say things like that. They came out because Joe had been thinking out loud for hours by then, with someone asking the next question and the next, until he reached formulations sharper and truer than the rehearsed ones. The extra sessions buy better thinking, caught the moment it’s said clearly for the first time, and a writer who skips them to save a few hours throws that away.
Why the quotes matter so much
I preserve quotes like these exactly as the client says them, because the phrasing is the voice.
Paraphrasing Joe’s line about the bike into something more polished would strip out everything that makes it work, the impatience, the bluntness, the dark humor that’s unmistakably his. A book in someone’s voice is built from hundreds of these preserved moments, the exact words they reached for when they were thinking hard instead of performing. Lose the phrasing and you lose the person.
This is also why the recording matters. Memory smooths things out. If I worked from notes, I’d remember the gist of what Joe said and lose the exact words, and the exact words are the point. The recording lets me capture the line as it was spoken, with the rhythm and the bluntness intact, so it can go into the book the way he said it instead of the way I would have tidied it up.
Incidental details get different treatment. When a client doesn’t remember or care what a minor character looked like, I make it up, and the client confirms it. I also swap real company names for made-up ones, because I don’t want my client to get sued.
Six sessions produced more usable material than the book will have room for, and that’s the goal.
When discovery ends with a surplus, the book gets built from the best of it, and it reads dense and confident. A book stretched to fill its length reads thin on every page. Ask any ghostwriter you’re considering how many hours of interviews they plan. If the answer is one long conversation, I’d keep looking, because you’d be paying for a book about your pitch.
Frequently Asked Questions
More than one good conversation. For Joe Rockey’s book, the equivalent of a full workday across multiple structured sessions. The value isn’t the volume of information but what happens to the author’s thinking when it’s examined repeatedly from different angles by someone whose only job is to understand it.
Because the first session is orientation, where the author gives the polished version they’ve delivered hundreds of times, optimized to end conversations instead of open them. The rough edges and interesting exceptions a book needs have been sanded off. It takes several sessions to get underneath the polish to the actual thinking.
The author stops presenting and starts examining. They follow ideas to new places, qualify positions mid-sentence, and return to stories they mentioned in passing. The sharpest, truest formulations tend to surface here, after hours of thinking out loud, often at the moment the author surprises themselves with a thought they hadn’t put that way before.
The specific phrasing is the voice, and polishing a blunt, funny line into something smoother strips out exactly what makes it land and what makes it unmistakably that person’s own way of talking. A book written in someone’s voice is built from hundreds of these preserved moments, the exact words they reached for when they were thinking hard instead of performing. Paraphrasing a sharp line into something tidier removes the impatience, the bluntness, or the dark humor that made it worth keeping in the first place. Working from a recording and not notes matters here too, because memory smooths things out while a recording captures the line as it was spoken, rhythm and bluntness intact.
Because memory smooths things out. Working from notes captures the gist and loses the exact words, and the exact words are the point. Recording preserves the line as spoken, with its rhythm and bluntness intact, so it can go into the book the way the author said it instead of tidied up.
Yes. Finishing with more strong material than you can use means the book gets built from the best of what exists instead of padded with everything that exists. A book assembled from surplus reads dense and confident, while a book stretched to fill its length reads thin, and readers feel that thinness on every page even if they can’t name why. Six structured sessions with a client like Joe Rockey produced exactly this kind of surplus, more usable quotes and material than the final manuscript will have room for.
