Privacy Policy

1. Introduction

At RetentIQ (hereafter "RetentIQ", "we", "us", or "our"), safeguarding your organizational telemetry and customer relationship data is one of our core architectural pillars. RetentIQ operates a state-of-the-art enterprise Customer Success (CS) and churn-intelligence monorepo platform.

This Privacy Policy explains how RetentIQ collects, processes, encrypts, and retains organizational profile data, active customer events, and transactional information. When you register a tenant account or connect integrations (such as Stripe, Intercom, Mixpanel) to our platform, you consent to the operations described herein.

2. Data We Collect

To predict qualitative customer churn indexes and construct health scores, our platform ingests data through direct API links, webhook dispatches, and manual imports:

• Customer Profile Metrics: Company names, contact emails, signup records, and monthly recurring revenue (MRR) figures associated with your customers.
• Behavioral Telemetry Events: User login counts, application feature adoption logs, and usage trends synced via tools like Mixpanel.
• Billing Information: Payment success histories, invoice failures, and contract renewal deadlines derived from your Stripe integration.
• Support Interactions: Ticket volume tallies, conversation timestamps, and qualitative sentiments synchronized via Intercom or other CRM platforms.
• Account Management Info: Workspace user details, email addresses, password hashes (handled securely via Supabase Auth), and workspace configuration variables.

3. Machine Learning & AI Processing

RetentIQ uses a hybrid intelligence engine consisting of an analytical machine learning module and a qualitative large language model (LLM):

• Local ML Classifier: We train a local LightGBM `LGBMClassifier` and compute SHAP (Shapley Additive exPlanations) values on the mathematical distribution of your telemetry properties (e.g. usage trends, support volumes, renewal proximity) to predict numerical churn probabilities. This training runs entirely within isolated compute boundaries.
• Groq AI Enrichment: To enrich numerical data with qualitative summaries, we send temporary, anonymized customer support ticket summaries and telemetry factors to the GROQ API (powered by Llama-3.3 models) to generate qualitative risk explanations and suggested playbooks. We enforce strict data policies with Groq to prevent your data from being retained or used to train public LLM models.

4. Third-Party Services & Integrations

RetentIQ does not sell or lease your customer success details to third parties. We transfer data to integrations only when authorized by you:

• Stripe, Mixpanel, and Intercom: We retrieve telemetry and payment histories under credentials you securely upload to your dashboard. This data remains strictly scoped within your workspace.
• Slack and SMTP Channels: Alerts and CS notification playbooks are dispatched to your Slack workspace or email channels (via SMTP configuration) based on thresholds you manage.

5. Multi-Tenant Data Isolation

RetentIQ is engineered on a secure multi-tenant architecture. We utilize Supabase PostgreSQL Row-Level Security (RLS) policies. Every table containing customer data, health scores, and playbooks is protected by strict policies that prevent users from other organizations from accessing or querying your data.

All data transfers are encrypted in transit via TLS 1.3, and database volumes are encrypted at rest using AES-256 standard encryption keys.

6. Your GDPR & CCPA Rights

If your organization is operating within the European Economic Area (EEA) or California, you are entitled to specific regulatory rights regarding your data:

• Right to Access: You can download or request reports of all telemetry records and metrics we store on behalf of your workspace.
• Right to Rectification: You can correct incomplete profiles or request database overrides.
• Right to Erasure (Right to be Forgotten): You can trigger a full deletion of your workspace database tenancy, which immediately wipes all customer events, health records, and AI summaries.
• Opt-out of Profiling: You can turn off automatic LLM-based playbooks and scoring parameters in your settings panel.

7. Data Retention & Deletion

We retain client telemetry and customer logs only as long as your workspace account is active. If a workspace remains inactive for more than 180 consecutive days or is explicitly closed, all associated tables, health histories, and configuration values are purged permanently from our production databases.

8. Contact Information

For questions regarding this Privacy Policy, compliance inquiries, or to execute a data deletion request, please reach out to our legal compliance group: