Analyze Shopping Data in Real-Time

ZOI CREATES A POS DATA LAKE ON AWS

DinoSol decided to migrate its applications, including a growing SAP environment, to Amazon Web Services (AWS) for increased scalability and flexibility. “AWS offers an excellent selection of cloud technology and services,” Prieto says. DinoSol engaged Zoi, a cloud-native technology services provider, and an AWS Premier Consulting Partner, for assistance. Zoi already managed all DinoSol SAP solutions, including the company’s business-critical enterprise resource planning (ERP) application.

Zoi moved the DinoSol PoS application and database into Amazon Relational Database Service (Amazon RDS) and containers through Amazon Elastic Kubernetes Service (Amazon EKS). Zoi then created an AWS-based PoS data lake, a centralized and automatically scalable data repository. In this new solution, an SAP Process Orchestration (PO) module analyzes purchasing information collected from each supermarket’s PoS system and sends it to Amazon API Gateway. From there, an AWS Lambda function extracts the data—including updated information on promotions, discounts, payments, and surveys—and sends the data to Amazon Redshift tables for visualization. 

At the same time, the solution stores plain XML documents in Amazon Simple Storage Service (Amazon S3) buckets and triggers Amazon Simple Notification Service (Amazon SNS). An SNS notification is captured by an AWS Lambda function that sends document data to the company’s Magento eCommerce application. Magento then assigns the data to a corresponding customer ticket. All DinoSol supermarkets are currently taking advantage of the PoS data lake.

We have to provide the strongest possible customer service, and we needed more scalability to meet the demand on our eCommerce system.

GAINING FULL INSIGHT INTO POS SALES DATA

DinoSol now has a centralized view of its PoS sales data thanks to the integration of the company’s PoS system, SAP ERP system, and eCommerce application. “The PoS data lake solves DinoSol’s challenge of decentralized PoS data,” says Juan Vargas, director of sales and marketing at Zoi. “As a result, all DinoSol departments can more easily access store transaction data in near-real time.”

Architecture diagram of PoS setup built on AWS.

GIVING SHOPPERS A BETTER EXPERIENCE

Because of its low-latency database and automated scalability, the PoS data lake can quickly ingest PoS data and grow to meet DinoSol’s business demands. Every day, about 150.000 PoS tickets are uploaded and stored in AWS from the moment of transaction. Each ticket is integrated from SAP PO into AWS in under 1,5 seconds. “We’re a growing company, and we can quickly add compute capacity on AWS to support thousands of new customers making purchases through our PoS system,” says Prieto. 

By uploading and accessing PoS data so rapidly, DinoSol can ultimately deliver an improved customer experience. “We want customers to immediately get a digital receipt on their mobile phone as they leave the store,” Prieto says. “We can do that with this solution by ingesting tickets as fast as possible, so we can help our stores run more smoothly and give our customers the best possible shopping experience.”

ANALYZING CUSTOMER SHOPPING PATTERNS

In the next phase of the PoS data lake implementation, DinoSol will use its PoS data to make better, faster decisions. Specifically, the company hopes to analyze customer purchasing patterns or the performance of specific seasonal products and to develop actionable insights in response. “For example, with the data in Amazon Redshift, DinoSol will be able to look at sales patterns to determine which products are top sellers,” says Vargas.

As it looks to expand its use of AWS, DinoSol is currently evaluating the development of a disaster recovery solution. “AWS is the right technology for our company,” says Prieto. “We look forward to working with Zoi as we continue our growth.”

ABOUT DINOSOL

DinoSol is one of the largest Spanish distribution companies and is based in the Canary Islands. They directly employ more than 7,000 people across 224 stores. Currently, the brand is present in five of the seven Canary Islands. Through the DinoSol Foundation, they support the social development of the Canary Islands and always pay special attention to the people who form part of the DinoSol Group.

GROWING CHAIN SEEKS SCALABILITY AND IMPROVED ANALYTICS

DinoSol Supermercados is a growing Spanish supermarket chain that operates more than 224 stores throughout the Canary Islands. The company has experienced significant growth over the past year with increasing demand on its digital eCommerce application, which thousands of customers use to purchase goods in stores and online. Because it hosted the application in an on-premises data center, the company struggled to keep pace with growth. “We have to provide the strongest possible customer service, and we needed more scalability to meet the demand on our eCommerce system,” says Eduardo Prieto, CIO at DinoSol Supermercados.

DinoSol also needed to centralize all sales information from each store’s point-of-sale (PoS) system so it could better analyze customer purchasing behavior. “We had decentralized PoS data and we wanted to find a solution that could enable us to collect it in one place and scale it quickly when necessary,” Prieto said.