Privacy Policy

Trust & Transparency: Grip Intelligence operates on a foundation of absolute transparency. We do not sell, rent, or trade your personal data. This is a non-commercial, personal tool; we do not use your personal information to train public AI models.

01. Data Collection & Usage

We collect information necessary to provide financial analytics, including your email address for account identification and transaction data extracted from bank-related notifications.

02. Google User Data (Restricted Scope)

Grip accesses your Gmail account using OAuth 2.0 Restricted Scopes (gmail.readonly). We strictly search for and process only bank transaction notifications, credit card alerts, and financial statements.

Restricted Scope Compliance: Our use and transfer of information received from Google APIs to any other app will adhere to the Google API Services User Data Policy, including the Limited Use requirements.

03. Transaction Extraction Engine

Grip uses a two-stage, privacy-first extraction pipeline to process bank notification emails:

Stage 1 — Local Rule Engine (Primary): Transaction details (amount, merchant, date) are extracted using a deterministic, in-process pattern-matching engine that runs entirely on our servers. No email content leaves our infrastructure during this stage. This handles >90% of all transactions with zero external API calls, zero latency, and zero cost.

Stage 2 — Groq AI (Fallback Only): For emails the rule engine cannot confidently parse, a sanitized and anonymized snippet is sent to Groq's external API. All PAN, Aadhaar, card numbers, UPI IDs, and phone numbers are masked before any data leaves our servers. Groq is used strictly as a last resort, not the default processing path.

04. Financial Intelligence Engine

Our logic-based intelligence engine performs rigorous financial calculations to provide variance analysis and "Safe to Spend" metrics. Safe-to-Spend is a deterministic mathematical calculation based on your current balance minus unpaid bills and projected recurring commitments. It includes a safety buffer based on your actual 30-day discretionary spending averages.

05. Privacy Shield & Sanitization

We implement Privacy-by-Design via a local sanitization layer that operates before any data is processed by our extraction engines or sent to any external service. Our system automatically detects and masks highly sensitive fields including:

This ensures that even during fallback external processing, your most sensitive identifiers are never exposed in raw format.

06. Security