Swiggy — Automating Payment Reconciliation at Unicorn Scale
A production-grade automation pipeline that replaced three days of manual finance work with a real-time reconciliation and data-lake analytics layer across every line of business.
What we were asked to solve.
Swiggy's finance team relied on an entirely manual cash-reconciliation process. Each cycle, three team members compiled data from Snowflake, Floating Cash Management Systems, bank statements, and third-party vendor feeds to produce reconciliation summaries covering total cash collected, payment-method breakdowns, cash movements, losses, and balances. Every line of business had its own SOP and cadence — SND and 3PL reconciled weekly, others monthly. Preparing a single report took three days, and the manual nature of the operation produced frequent errors and downstream losses. As Swiggy scaled, the process could not scale with it.
How we engineered the solution.
SOP-to-Workflow Translation
Codified every line-of-business SOP into automated workflows using Microsoft Power BI and Power Automate — capturing the operational logic finance had been running manually for years.
Multi-Source Data Ingestion
Built an ingestion layer pulling data from Snowflake, FCMS, bank statements, third-party vendors, and Google Drive — normalizing everything into a consistent reconciliation model.
Data Lake Analysis
Structured the aggregated finance data into a lake-style analytical layer, enabling variance detection, cross-LOB queries, and drill-down analysis that were previously impossible on the raw sources.
Automated Reconciliation & Variance Detection
Each workflow runs the SOP end-to-end — fetch, transform, reconcile, and flag variances — with no human intervention required for the routine path.
Real-Time Reporting Layer
22 different reports generate in real time — from total payments processed and incomplete transactions to payment dues and cash movements — with a holistic executive view across every LOB.
Stakeholder Distribution
Reports surface directly in Power BI for finance stakeholders, or auto-export and distribute via email for teams that prefer scheduled delivery.
What changed for the business.
90 hours saved per month for the finance team — the equivalent of a full engineer's headcount.
22 different reconciliation reports generated in real time across every line of business.
Reduced turnaround time from three days per cycle to on-demand.
Dramatically improved accuracy — automation removed the manual entry errors that were causing downstream losses.
Enhanced visibility for finance leadership across cash collections, movements, losses, and balances.
A compliance-ready audit trail with data lineage back to source systems.
A scalable analytics foundation that grows with Swiggy's LOB expansion.
Other case studies
AI-Powered Sales & Content Technology
Logistics & Freight Technology Platform
IoT Monitoring & Operational Visibility
ERP Implementation for Gallop Brush Manufacturing
Shipolog — AI-Powered Fulfillment Platform for E-Commerce Sellers
Facing a Similar Engineering Challenge?
Bring us the context of your product, workflow, or technical constraint — we will help identify the right path forward.