Legal
Subprocessors
These are the third parties that process data on our behalf to run Churnless, what we use each one for, and what data it receives.
Last updated: September 22, 2026
Current subprocessors
| Provider | Purpose | Data processed |
|---|---|---|
| Vercel | Application hosting, serverless functions, background workflows, agent sandboxes, and private object storage (Vercel Blob) for session replay chunks. | All data processed by the application, including replay chunks stored for 30 days. |
| Neon | Managed Postgres database, the system of record for accounts and customer data. | Account, workspace, customer, subscription, cancellation, feedback, and recovery records. |
| Tinybird | Analytics database for product events and dashboard metrics (AWS us-west-2). | Behavioral, billing, cancel-flow, and recovery events with customer identifiers. |
| Anthropic | AI models (Claude) for feedback analysis, health insights, recovery planning, and drafting retention emails. | Customer context such as name or email, subscription details, cancellation reasons, feedback, email replies, and text summaries of privacy-masked session replays. |
| Vercel AI Gateway | Routes some AI model requests to the model providers listed here. | The same model inputs as the provider the request is routed to. |
| TypeSafe (Jev) | Bounded AI judgments, such as choosing which configured cancel-flow offer fits a cancellation and checking agent decisions against policy.Only when enabled by Churnless. Requests go through Vercel AI Gateway. | Product and plan details, cancellation survey answers and free text, and distilled customer facts. |
| Resend | Sending transactional, dunning, and retention emails, and receiving replies to them. | Recipient name and email address, message content, and reply content. |
| Stripe | Billing for Churnless subscriptions, and reading and updating your own Stripe account when you connect it. | Churnless billing details; for connected accounts, customers, subscriptions, invoices, and charges. |
| supermemory.ai | Relevance-ranked retrieval of distilled customer facts for AI agents.Only when configured by Churnless. | Short facts scoped per organization. Anything containing an email address, phone number, or card-like number is rejected before sending. |
| context.dev | Looking up public brand details (logo, colors) for your company during onboarding.Only when you use the brand lookup step. | Your company's website domain. |
| Upstash | Shared rate-limit counters for public API and browser endpoints.Only when configured by Churnless. | Rate-limit keys made of API key IDs, organization IDs, and IP addresses, with request counts. Entries expire with each limit window. |
| Sentry | Error reporting for server-side failures.Only when configured by Churnless. | Error-level application log events, such as the request path, error details, and deployment metadata. |
| PostHog | Product analytics for Churnless's own signup and onboarding funnel.Only when configured by Churnless. | Churnless account users' funnel events and campaign attribution. Not your customers' data. |
Integrations you connect
Services you connect yourself, such as your own Stripe account or a Slack workspace for notifications, receive data because you configured them. They are governed by your agreement with that provider and are not listed as Churnless subprocessors.
Regions and changes
Contact us to confirm the hosting region for each provider. We will update this page when we add or replace a subprocessor. For questions, or to request a DPA, email privacy@getchurnless.com. See also our Privacy Policy and Security page.