The client is an American chain of fast-casual restaurants offering chicken wings, chicken fingers, sandwiches, and salads. They wanted insights into operations, sales, and customer service at 900+ locations to make informed business decisions. They wanted to monitor store performance, the impact of special offers on sales, and identify the mix of orders coming through walk-ins, online, drive-thru, etc. These insights would help them manage inventory, employee distribution, and maximize the impact of offers. To enhance the experience of their most loyal customers, the client also wanted to create a holistic view of customer behavior by incorporating loyalty and digital data.
Dealing with siloed analytics data from legacy on-premises systems and tools restricted the client’s ability to accelerate the massive digital transformation the brand planned to undertake. To drive success, the team knew they must break down silos, connect datasets, and gain efficient reporting with a scalable and secured solution, having controlled definition of business metrics to get a more holistic and actionable view of common questions (sales, cost, labor, etc.).
The client turned to Persistent to help build a scalable and modernized Data Lake and Warehouse on the Google Cloud Platform (GCP). The platform would democratize data by providing universal and secure access while creating a business intelligence self-serve reporting mechanism for a comprehensive and timely information ecosystem.
Persistent develops a data platform using Google Cloud Platform
Persistent managed the end-to-end modernization of the client’s on-premises data warehouse to the Google Cloud Platform (GCP), creating a data lake on Google Cloud Storage and a data warehouse on BigQuery. Our data experts worked with the client to define data attributes and features, such as the wait time at the drive-thru, rate of order cancellations, labor productivity, historical sales, average check size, percentage of online orders, and percentage of a strategic category. The Persistent team pulled huge volumes of data from various input sources, automating the ingestion process, complex cleansing and transformation routines, before making the data available for analysis. The team also encoded business rules in Looker, such as tallying the car count with the drive-thru check count, stores with no revenue for the day, etc. thereby alerting for possible business rule violations.
We created fault-tolerant data pipelines by building reusable components with Google’s serverless offerings that reduced management and infrastructure overhead. This data platform helped the client with a single source of truth for factors impacting sales across locations, which, when fed into Looker, helped the business users create custom dashboards without external help.
The data sourcing, pipeline creation, and data transformation were governed by GCP best practices, and we set up everything in the environment programmatically using recommended Continuous Integration and Deployment (CI/CD) via Terraform and scripting languages.
A single source of truth fosters data availability, accuracy, and trust
Persistent’s Google-backed data platform analyzes 45 million point-of-sale transactions daily, providing business leaders with a detailed view of sales reports, along with drill-down capabilities for further insights. It also sends leadership alerts when pre-defined KPIs are met, helping the client measure business goals against real-time data.
Built on the powerful GCP stack, the data platform reduces operating costs by 60% by eliminating licensing costs and providing intelligent data storage and computing costs on the customized data model.
This platform enables business users to analyze data and create dashboards on their own, eliminating the need for developers. The client’s data team doesn’t have to manage the platform, freeing them to focus on expanding their AI and machine learning capabilities. This has empowered data scientists and analysts to spend more time analyzing data and less time requesting data access from gatekeepers.
