The craft
What Should a Memoir App Actually Do for the Storyteller?
A memoir app is really six separate capabilities — conversation, memory, review, export, privacy, and writing burden. Here's how to judge each one.

A memoir app's marketing page usually shows the same promise: answer some questions, and the app writes your life story back to you. That promise hides the actual product, because "memoir app" isn't one feature — it's a bundle of separate capabilities, and two apps that make the identical promise can differ completely in how they handle conversation, memory, review, export, privacy, and the plain question of who does the writing.
A memoir app is worth judging on six capabilities, not its pitch: whether it holds a real conversation or just records answers to fixed questions, whether anything it learns carries across sessions, whether the storyteller reviews and corrects the output before it counts as final, what you can take with you if you stop using it, who can see or reuse what you said, and how much of the actual writing is still left for you to do. A fluent demo answers none of these on its own.
What a memoir app's promise usually leaves out
The search results for "memoir app" mostly show one shape of product: a chat interface or a form, a set of prompts, and an AI-generated narrative assembled from the answers. The pitch is speed — a life story without the blank page. What the pitch rarely specifies is what happens to an answer after you give it. Does a vague reply get a follow-up, or does the app move to the next question regardless? Does the app remember what you said last week, or does every session start from nothing? Can you see and change the paragraph it wrote about you, or does it publish what it generated as-is?
None of that shows up in a screenshot. It only shows up in use, which is exactly why it's worth naming the capabilities before you try the product rather than after you've already trusted it with a story.
Six capabilities worth judging in a memoir app
Conversation. A form with fixed questions and a chat interface that responds to what you actually said are different products wearing the same interface. The tell is the follow-up: does the app ask a second question shaped by your first answer, or does it move down a script regardless of what you said? A short, general answer — "it was a normal childhood" — is where a real conversation does its work, and where a scripted one has nothing left to offer.
Memory. Some memoir apps treat every session as a fresh start; nothing from last week's conversation shapes this week's questions. Others build a running record, so a name or a place mentioned once can be asked about again later, in more detail. A memoir app with no memory across sessions is really a series of unconnected interviews, however continuous the interface makes it feel.
Review. This is the one that matters most, and the one marketing pages say the least about. Before an app's version of a story counts as finished, does the storyteller see it and get to correct it — or does the generated text simply get delivered as the final product? The Oral History Association's published practice guidelines describe narrator review as a standard part of a responsible process: the person who told the story gets to read the account back and fix what's wrong before anyone treats it as settled. A memoir app that skips that step isn't a faster version of the same thing; it's a different, riskier thing that happens to look similar.
Export. If you stop using the app in six months, what do you actually keep — the raw recordings, a transcript, the finished narrative, or nothing beyond what's on the company's servers? A product's answer to "what if I leave" is a better predictor of how it treats your material day to day than anything in its onboarding flow.
Privacy. Who can see what you said — the company's staff, a training pipeline for the same AI model, other users, nobody but you? A memoir app that fabricates a voice reading, a video, or a quote in someone's likeness without their consent has crossed a real line, not a cosmetic one; recording a conversation is not permission to copy someone's voice covers that boundary directly, and the same principle applies to any generated likeness, not only audio.
Writing burden. This is the honest question underneath all the others: after using the app, who actually wrote the sentences — the storyteller, the app's model working from a transcript, or some blend where it's no longer clear which parts are whose words? A tool that fills in details a person never said, to make a paragraph read more smoothly, isn't saving anyone writing effort — it's inventing content and presenting it as memory. A memoir shouldn't promise a memory it never received, and a family should be able to trust that an uncertain or disputed detail stayed uncertain in the final account rather than getting quietly resolved for readability.
A memoir app checklist you can use on any product
Before trusting a memoir app with a real story, ask it — or its terms of service — these seven questions:
- Does it ask a follow-up based on what I actually said, or only the questions it had prepared in advance?
- Does anything I say in one session change what it asks me in the next one?
- Do I see and approve the finished writing before it's treated as final, or only after?
- If I misspoke, changed my mind, or want something removed, is there a real way to fix it — or is the record treated as permanent the moment it's generated?
- What do I keep if I stop using it — recordings, transcript, finished writing, all three, or none?
- Who can see or reuse what I said, and does that include training a model?
- Is it clear, in the finished text, which parts came from me and which parts the app filled in?
A product that answers all seven plainly is being straightforward about what it is. Vague or evasive answers to the review and privacy questions are worth treating as a real signal, not an oversight — the same category error that made "memoir app" sound like one simple thing is often what a vague answer is quietly relying on.
Where Joan fits, honestly
Joan is a guided conversation, not a form. Calls happen over time rather than in one sitting, and a detail mentioned in an earlier call can shape a later question rather than disappearing into a transcript nobody revisits. Nothing a storyteller says becomes part of the archive until they've reviewed and approved it — nothing enters the archive without approval explains that process in full rather than summarizing it again here. Joan doesn't invent detail to smooth a paragraph, and doesn't generate a voice or likeness beyond what was actually recorded.
That doesn't make Joan the right fit for everyone measuring a memoir app against this list — someone who wants to do all the writing themselves, on their own schedule, with no ongoing conversation, is better served by a prompt book or a private journal. But the six capabilities above apply to any product wearing this label, Joan included, and they're worth checking before the story you're trying to preserve is the one on the line.
If you'd rather see how a guided conversation actually works than take an app's word for it, the first call is free and there's no card.
