Signing in with Google
Narrative Lab accounts use Google sign-in only, and we never see your Google password. From Google we keep your name, your email address and whether Google verified it, a link to your profile picture, and the identifier of your Google account. Google’s access tokens are not kept.
Each sign-in creates a session. It records the IP address and browser description (user agent) it was started from, and it sets a cookie that keeps you signed in.
The books you create
For each book we store your brief, the record that links the generation to your account, and everything produced. The brief includes your topic and choices, the creator name you entered if any, and whether the book is public or private. What is produced includes the book, its cover, its research notes, and production details such as the models used and their cost.
Public books can be read by anyone. Private books are not listed publicly: you can open your own when signed in, and Narrative Lab’s administrators, and anyone Narrative Lab gives private reading access to, can open them too. Administrators can also see each account’s name, email address, books and credit balance.
When you use “Explore this topic”, your topic and the choices in your draft are sent to OpenAI for suggestions. We keep only usage figures from it (counts, timing, cost), not the topic.
Credits and payments
We keep a record of your credits. It holds every credit you receive, each book’s hold and whether it was spent or returned, and the promotional codes you redeem.
Payments are handled by Stripe on its own checkout page. Stripe collects your card details and email address, and we never receive your card number. Stripe receives your Narrative Lab account identifier with the purchase. For each purchase we keep the Stripe checkout and payment identifiers, the pack, the number of credits, the amount, the currency and the date. Stripe’s own privacy policy applies to what it collects.
Services that process your data
- OpenAI writes the books. Your brief, and the text being written, are sent to OpenAI’s models to write the book, check facts (including by web search) and create the cover.
- ElevenLabs narrates audiobooks. When an audiobook is made of a book, its text is sent to ElevenLabs. Only Narrative Lab’s administrators make audiobooks.
- Google provides sign-in. The site’s fonts are loaded from Google Fonts, so your browser connects to Google when it displays a page. The optional reader survey at the end of a book is a Google Form.
- Stripe processes payments.
- Wikimedia Commons and other public archives: while a book is being written, search terms taken from it may be sent to Wikimedia Commons to find research images. When a book shows such images, your browser loads them from the archive that hosts them.
- Netlify (the site and its file storage), Neon (the database) and Railway (book generation) host the service. They store or process the data above to run it, along with technical data about each request, such as IP addresses.
Each of these services handles data under its own terms and privacy policy.
Reading statistics
When you read a book, the page sends reading events. Each event records the book and chapter, your progress and reading time, the audio position and speed, and your answer to the end-of-book question. The events carry a random identifier stored in your browser, not your account. A book idea sent through the form at the end of a book is stored the same way. Narrative Lab’s administrators, and anyone it gives private reading access to, can see these statistics and ideas.
We also keep usage figures for book creation and Explore (counts, timing, cost) to run the service.
Cookies and browser storage
Narrative Lab sets a few cookies:
- one that keeps you signed in;
- one that secures a sign-in while it is in progress;
- for people given private reading access, one that remembers it.
Narrative Lab sets no advertising or third-party analytics cookies. Reading statistics use a random identifier kept in browser storage, described below.
Your browser also keeps some data on your device: your unfinished book draft, the interface language, your reading progress and audio preferences, and the random reading identifier. Clearing this site’s data in your browser removes them.
Keeping and deleting data
We have not set fixed retention periods, and data is not deleted automatically on a schedule. There is no self-service way to delete an account yet. You can write to support@narrative-lab.app to ask what we hold about you and what can be deleted. These requests are handled by hand.
Some records cannot be deleted. Once an account has created a book, or has received, used or bought credits, its account record is kept with the records of those books, credits and payments, because credits and purchases cannot be accounted for without them. A book can be removed from the site, but the record of its generation is kept. Stripe keeps its own records of payments.
Changes and contact
We may update this page; the date at the top shows when it last changed. Questions about your data: support@narrative-lab.app.