Privacy

Last updated 2026-05-31

This page exists because two claims on our homepage are real promises, not marketing copy: we don't train on your data, and we don't store your bills. The rest of this page is the plain-language version of how we deliver on those, and what we do collect when you actually use the product.

What we collect

Your email address when you join the waitlist or create an account. We use it to send you product updates and account notifications. We do not sell it, share it with third parties, or use it for cross-site advertising.

The bill image when you upload one to be read. The image is held in memory only for the seconds it takes to extract the data. It is not written to disk, not retained in any database, and not used to improve any model. The structured output we return to you is yours to keep; the source image is not retained.

Basic operational data when you visit the site or use the product: page views, error logs, your subscription state if you are a paying user. We use this to keep the service running and to fix bugs. We do not run third-party advertising trackers.

What we do not do

1. We do not train AI models on your data. The extraction model we use is a paid-tier API call to a foundation-model provider. Under the paid tier's terms, your inputs are not used to improve the provider's models. If those terms change, we will say so on this page within a week of the change.
2. We do not retain your bills after processing. Bills are streamed through the extraction pipeline and dropped. There is no backup, no archive, and no analyst at our end reviewing the documents you sent. The structured data we return to you is the only persistent artifact, and it is attached to your account, not to a training set.
3. We do not sell, rent or share your data with third-party marketers or advertisers. Your email and account data is used only to operate MessyDocs.
4. We do not use dark patterns to make cancellation harder. If you cancel, your account is closed and your subscription stops at the end of the current billing cycle. Email us if you want your account deleted entirely.

Sub-processors we use

We use a small number of external services to actually run MessyDocs. Each one sees only the slice of data needed to do its job.

• The foundation-model provider processes the bill image to extract structured data. We send the image; we do not send your email, subscription state or any other account data. The provider's paid-tier terms prohibit using your input for model training.
• Our authentication and database provider stores your email, account state and subscription status. They do not see the bills you upload.
• Our payment provider handles subscription billing. They see the payment information you give them at checkout (UPI mandate, card details). We never see or store your card numbers; we only see a token confirming the subscription is active.
• Our hosting provider serves the site and the API. They see the network requests but not the bill content (which is processed in memory and never stored).

How long we keep things

• Bill images: not stored. Held in memory during processing, dropped immediately after.
• Structured output (the data we extract): kept while your account is active so you can download it. Deleted when you delete the row or close your account.
• Email + account data: kept while your account is active. Email us to delete entirely.
• Payment records: kept for the duration required by tax law (typically 7 years), then deleted.
• Server logs: kept for 30 days, then rotated.

Your rights

You can request a copy of the data we hold about you, ask us to correct anything wrong, or ask us to delete your account and the associated data. The fastest way is to join the waitlist for now (we aren't live yet); once we are live, the same request can be sent over email. We answer within a reasonable time, typically within a few business days.

If you are in a jurisdiction with a specific data-protection law (DPDP in India, GDPR in the EU, UK GDPR in the UK, similar elsewhere) those rights apply to you the same way they would for any other small online service. We will respond to lawful requests from regulators in your jurisdiction.

Changes to this page

If we change anything material on this page, we will update the "Last updated" date at the top and, if you have an account with us, email you about the change. Cosmetic edits (typos, clarifications) we just make.

Questions

Send them via the waitlist for now; we are not live yet. Once we are, the same questions can be sent over email and we will answer them.