Prosed
Go from newsletters & podcasts to publish-ready book
246 followers
Go from newsletters & podcasts to publish-ready book
246 followers
You've spent years creating—newsletters, podcasts, LinkedIn posts, courses. The book is in there. You just haven't had time to assemble it. Prosed's Inkwell pipeline analyzes your voice, structures your scattered content into chapters, and produces a manuscript that actually sounds like you. Not generic AI writing or slop. Your words, your ideas, assembled into something real. Built-in editorial review. Print-ready PDF/DOCX export. Beta: $47 for the first 100 founders.
This is the 2nd launch from Prosed. View more
Prosed
Launching today
Most AI book services work like a black box. Your content goes in, a manuscript comes out, and you hope it sounds like you.
We thought creators deserved more than hope.
The Source Map color-codes every paragraph in your manuscript: your original words from newsletters, podcasts, and posts alongside the connective tissue Prosed added to bridge your ideas together.
Today, over 90% of each book we assembled is the creator's own words. Now you can see that for yourself.
Payment Required
Launch Team / Built With
Prosed
Congratulations on the launch @rails
I'm really interested in how the "your voice" matching holds up across formats. I found that my newsletter prose, podcast transcripts, and LinkedIn posts each have a completely different register, sentence length, punctuation habits. What's actually normalizing those into one consistent narrator voice, and does the writer ever get a say in resolving the conflict when the sources disagree?
Prosed
@aidan_codefox Hey Aidan, thank you, and what a good question :)
You're right that your newsletter self, your podcast self, and your LinkedIn self are three different writers. We see it in almost every archive.
Two things do the normalizing:
First is the creative brief, where you tell us directly how you want the book to sound: the phrases you love, the ones you never want to see, the writing you consider your best. That brief is the referee when your sources disagree
Second is that written and spoken material do different jobs in assembly. Your written prose tends to set the narrator's rhythm and register, while transcripts contribute stories, ideas, and the way you explain things out loud.
And yes, you absolutely get a say. Every chapter arrives in your portal for review, and if the narrator doesn't sound like the you that you wanted, you can annotate it, edit it directly, or tell us which source's voice should win. The book should read like your best writing self, and creators are the judge of who that is!
Interesting that the second launch is about proving the first one. Most teams would have shipped more features and this one shipped receipts instead. Says something about who your buyer is. Question: after the Source Map, is the remaining objection still trust, or has it moved to something else like whether the book is actually good?
Prosed
@arya012 Hey Arya, "Shipped receipts instead of features" might be the best one-line summary of this launch I've read, including our own 🧡
And your instinct is right. Once someone sees the map, the question changes from "is this really mine?" to "is it actually good?" Honestly, we're glad it does, because that's the question a publisher should be judged on. It's why every chapter gets scored against markers we modeled from bestselling nonfiction, and why an editorial reviewer checks voice, flow, and cohesion before a creator ever approves a book.
But "good" is subjective, and no scoring system settles it on its own. That's why every book that goes through Prosed is also read by a teammate of ours who is a well-versed writer herself. A human who reads it the way a reader would, before it ever reaches the creator. So when we say the good part is there, we're not guessing! :)
Might this be the place where humans will return? 🤓
Prosed
@thys_beesman Hey! Thank you for stopping by. Elise, co-founder of Prosed. This is exactly the kind of question we hoped today would surface, so thank you for taking the time.
On your first point, you're right, and I want to be precise about what the map claims. Source Map shows provenance: which words came from your original material and which ones we added. It does not measure how much the surrounding structure changed the emphasis of your words, and I won't pretend it does. Word share and meaning share are different things.
What guards the meaning side is what happens around the map. We don't invent arguments, and the framing follows the creative brief where you tell us what the book is arguing and how it should sound. After assembly, our editorial reviewer scores every chapter on voice fidelity, story flow, and cohesion, specifically to catch places where structure drifted from what you were actually saying. And nothing gets approved except by the creator, who reads every chapter. Is that the same as a meaning indicator on each paragraph? No. But it's an idea worth sitting with, and I've written it down.
Your second question has a happier answer. You can edit any bridge directly in the portal. Select the paragraph, rewrite it or remove it, and it's done. No need to go back through the whole process. A full regeneration exists for when you want your edits woven through the entire book, but pulling apart a seam you don't like is a direct edit, right there in the chapter.
And your last line is the standard we're building against!
Turning existing content into a book is useful, but the 'shows its work' part is what makes it credible. Can authors reject a weak source or paragraph-level claim before the manuscript gets assembled, or is the evidence mainly there for review afterward?
Prosed
@yaroslav_stelmakh Hey Yaroslav, thank you for this one! The curation happens at two moments, and only one of them is before assembly.
Before we assemble anything, you choose what we work from. You decide which newsletters, transcripts, and posts to share; you can label each batch with how we should use it, and the creative brief lets you tell us what to lean on and what to leave alone. So a weak source can be kept out entirely, or fenced in as reference only. Up to you/the creator.
Paragraph-level rejection happens after, in review. Every chapter comes to your portal, and anything that doesn't belong can be edited directly or cut. I'd love to get to a place where you can approve claims before assembly too, but that isn't the case today. What exists is a fast loop for pulling out anything that doesn't resonate for the creator. But if it's there, it's the input it was given. The pipeline is actually really good. See yourself? 😎
Have users become more comfortable publishing once they can actually see what the AI contributed?
Prosed
@reda_roqai_chaoui Hey Reda! Honest answer: the map went live for everyone today, so I'll have better data in a month than I have this morning. But I can tell you what we saw with the creators who had early access.
The hesitation was never really about quality. It was a quieter worry, that they'd say "an AI wrote your book." And the first thing nearly every creator did with the map was toggle it on and just scroll, looking at all that green and understanding the % split.
The shift I find most interesting is what they do next. They show it to people. The map turns "I hope nobody asks about the AI" into "let me show you exactly how this was made."
Ask me again in a month, and I'll tell you whether better data backs up what our beta creators and I are seeing!
Prosed
One more thing we made for today 📖
The question behind Source Map is "how much of this book is really mine?" So we built a little page that answers it before you even start.
Count what you've already published: newsletters, podcast episodes, blog posts, videos, course modules. It adds up the words and shows you how much of your book already exists.
Most creators are shocked. Someone with 100 newsletters and 50 podcast episodes is sitting on 8.8 books' worth of material 🤯. Not 1 book. Almost 9.
Try it with your own numbers: https://tryprosed.com/creators
And if your archive turns out to be a library, well. You know where we are today. 🧡