Skip to content
Back to Work
Chick-fil-A - Offline-First Ordering & Device Mesh logo
Enterprise Solution

Chick-fil-A - Offline-First Ordering & Device Mesh

Completed: 2020

Architected an offline-first ordering system using Ditto and event queues for seamless device-to-device communication — empowering festivals, food trucks, and food carts while migrating off legacy Aloha POS.

The Problem

Chick-fil-A needed to support diverse ordering scenarios beyond traditional in-store: festivals, food carts, food trucks, and pop-up locations where WiFi is unreliable or nonexistent. The legacy Aloha POS system couldn't handle offline scenarios, and when WiFi dropped or internet degraded, orders were lost or delayed — unacceptable for a brand built on speed and excellence. Device-to-device communication was fragile, and there was no resilient way to synchronize orders across tablets, kitchen displays, and payment terminals when connectivity failed.

The Solution

Served as Sr. Solution Architect through EY on a $20M engagement. Architected an offline-first ordering system using Ditto for peer-to-peer data synchronization and event queues for reliable message delivery. Ditto enabled device-to-device crosstalk in case of WiFi loss, bad internet, or other connectivity issues — tablets, kitchen displays, and payment terminals could continue operating and communicating with each other without any cloud dependency. Implemented a fully integrated event system that allowed fast updates, device-to-device communication, and eventual rectification with the web once connectivity restored. Orders were processed and parsed locally, then synced upstream as connections became available. This architecture enabled migration off the legacy Aloha system while supporting diverse ordering scenarios.

The Impact

A $20M EY engagement to modernize ordering at Chick-fil-A off the legacy Aloha POS. The offline-first architecture was aimed at preventing lost orders when WiFi or internet dropped — device mesh via Ditto keeping kitchen displays, tablets, and payment terminals in sync without a cloud connection, with an event queue to reconcile with the backend once connectivity returned — designed to hold up across 2,600+ locations plus festivals, food trucks, and pop-ups. The larger goal: get them off the locked-down, proprietary Aloha system — unmodifiable, with painfully slow turnarounds — and onto a modifiable architecture that would unlock much faster development. The engagement was still underway, so these were the targeted outcomes, not finished results.

Technologies

Ditto
Event Queues
Microservices
Offline-First
Device Mesh
Solution Architecture

Share This Case Study