Security and incident response
How Hailbell protects the data your apps and stores send us, and exactly what we do if that data is ever exposed, lost or reached by someone who shouldn't have it. Last updated 15 November 2026.
How we protect data
- Where it lives. Hailbell runs on Cloudflare. Its database is encrypted at rest, and every connection to Hailbell uses HTTPS.
- Keys get a second lock. Push keys (Apple and Firebase) and Shopify access tokens are encrypted again with a separate key before they're stored, and the dashboard never shows them after upload.
- Each account sees only its own data. Every dashboard, API, MCP and webhook request is checked against the account it belongs to. Inside an account, owners, admins and agents get different rights.
- Sign-in. People sign in with Google or a one-time emailed link, so Hailbell stores no passwords. Sessions can be ended from the dashboard. The accounts that run Hailbell itself (Cloudflare, GitHub, Google, Shopify) use two-factor sign-in.
- Logs. Each account has an audit log of the changes made in it and a log of who opened which customer's chat, details or orders, kept for 12 months. Cloudflare logs requests to Hailbell's servers.
- Backups. The database can be restored to any minute in the last 7 days. Test data and production data are kept apart: tests use made-up data and a local database.
- Only what's needed. We keep data only as long as the terms say. For example, chats are deleted 12 months after their last message, and Shopify orders are read when a chat is opened and never stored.
If something goes wrong
A security incident is anything that exposes, changes, deletes or gives someone access to data they shouldn't have. That includes a leaked key or token, someone reaching another account's data, data lost beyond what a backup restores, or a bug that shows one customer's data to another. Vidur Saini, Hailbell's founder, is responsible for handling every incident.
What we do, step by step
- Look into it right away. Every report and alert gets a first look within 24 hours. We decide whether it's an incident and how serious it is.
- Stop it. We cut off the access: rotate or revoke the leaked key, token or secret, end the affected sessions, turn off the broken feature, or pause the affected account if we have to.
- Work out what happened. We find out which data was involved, which accounts, apps and stores were affected, when it started, and how.
- Tell the people affected. See below.
- Fix the cause. We fix the bug or gap that allowed it, check the fix, and restore any lost data from backup where we can.
- Learn from it. Within 14 days we write down what happened, what we changed and what we'll do differently, and we share that with the affected account owners.
Who we tell, and when
- Account owners. We email the owner of every affected account within 72 hours of confirming the incident. The email says what happened, what data was involved, what we've done, and anything they should do, such as telling their own customers.
- Shopify stores. When the incident involves data from a Shopify store, we tell that store the same way, and we tell Shopify as our partner agreement requires.
- Regulators. Where the law requires it, we notify the relevant authority within its deadline.
If we learn more after the first email, we send an update.
Report a security problem
If you think you've found a security problem in Hailbell, email hello@hailbell.com with what you found and how to reproduce it. A person reads every message, and we reply within 2 business days. Please don't access or change other people's data while testing, and give us a reasonable time to fix the problem before you share it publicly.
We review this policy at least once a year and whenever Hailbell changes in a way that affects it.
See also: Privacy, Your apps' data and Subprocessors.