Best known for its gift bouquets made of juicy, fresh fruit, Edible Arrangements (Edible) started as a store front in East Haven, Connecticut, at the turn of the century. Now, Edible serves 100 million customers—50 million gift givers and 50 million gift recipients—around the world, having branched out from bouquets to other sweet and chocolatey gifts.
With more than 2,000 brick-and-mortar stores and an e-commerce platform, a typical day can have Edible’s datacenter processing an average of 50,000 orders. On a gift-giving holiday, that number can skyrocket to between 800,000 and 1 million orders. As the business and product catalog has expanded over the last decade, Edible has been moving its business processes into the public cloud. To maximize the value of its data and make data-informed decisions, Edible has collaborated with Microsoft to build a new data warehouse with Azure Synapse Analytics.
A growing business needs room to scale
Siloed data impacted the accuracy and speed of reporting, which could only happen on an ad hoc basis through a ticketing system. A department or team would submit a report request to a team whose job it was to manually locate data, analyze it, and create a report. Because data was siloed, the reporting team didn’t always return results in the same terms that the requesting team needed.
“Each department may come up with their own numbers,” Ilango Kasiraja, Senior Data Architect at Edible says. “That was one of the key problems we wanted to solve by bringing the data warehouse in, where we consolidate all the information in one place. And all the reporting goes out of that, so that any department who wants access will always look at the same view of the data.”
“Performance was key, as well, because we have SLAs to meet,” says Kasiraja. “All the sales reporting and year-end and month-end reporting has to finish on time. Every day by 6:00 AM, we have to open up the shop for the executives to look at the sales numbers for the previous day.”
“Post-COVID, consumer behaviors have changed drastically. People are looking for a last minute, last mile, so the time of delivery, how soon, and how fast—those are very important consumer questions that we’re trying to answer as a business right now.”
Ilango Kasiraja, Senior Data Architect, Edible Arrangements
Using data storage and analytics to accelerate time to insight
A data warehouse and analytics platform in a public cloud could give Edible the scalability, agility, insights, and cost effectiveness that it needed to accommodate the massive amount of orders it gets on gift-giving holidays. Exploring Azure Synapse Analytics was a natural choice for Sai Padmanaban, Vice President of Technology at Edible, because it combines data integration, enterprise data warehousing, big data analytics, and a flexible querying system.
Azure Synapse gives Edible insight into its operations, supply chain, inventory demands, and retail operations, leading to data-informed decisions and efficiencies. And interoperation with Microsoft Power BI and machine learning helps store and district managers mitigate inventory shortages. If there’s no cheesecake coming in, then that affects the store’s ability to make products that incorporate cheesecake—a shortage that could affect dozens of confectionary gifts. On the flip side, this insight can help stores reduce food and ingredient waste.
Edible’s technology team wanted to do its due diligence, so it evaluated and compared Azure Synapse and two third-party data warehouse solutions. Azure Synapse matched or exceeded the cost-to-performance ratio, and it would be easier and faster to deploy.
“We had a timeline of four to five months,” Kasiraja says. “The tools and services Azure offers helped with a quicker to-market data warehouse.”
Azure Synapse performance, cost, and ease of implementation was a better deal and made the tight deadline achievable.
“So far, I think we made the right decision by going to the cloud. With Azure, we not only lessen the stress of maintaining and managing our infrastructure, but we can also better focus on business transformation and accelerate our speed to market on our digital, data, machine learning, and cognitive-related projects.”
Sai Padmanaban, Vice President of Technology, Edible Arrangements
Microsoft manages the cloud infrastructure, enabling Edible to dedicate those resources to transforming its business. Knowing that a Microsoft team is tasked with ensuring high availability and high performance of the cloud systems puts Padmanaban at ease.
Edible’s Azure-based architecture
Kasiraja says, “We have a lot of data to crunch—customer data, order data, product data—and everything has to go through the ETL [Extract, Transform, Load]. We have data lakes in between and multiple hops for the data to make before it can land in the data warehouse, then data mart, and finally, the data set in the report has to be refreshed in time.”
Data streams from the datacenter to the integration and storage layer where Azure Data Factory, Azure Data Lake, and Azure Databricks sit. The data factories hook up to Azure Synapse, which lives in the analytics and semantic layer of the stack. Azure Synapse stores processed and ready-to-use customer insights data and includes an analytical layer that provides rapid reporting via Power BI. The team also starts to take advantage of Azure Purview in preview to help the company automatically scan and classify data across its data estate.
Without having to install anything, Edible was able to use Azure Cosmos DB and an instance of Azure Database for PostgreSQL to launch new business entities and initiatives. Edible has also connected Azure Synapse with Microsoft Dynamics 365 Customer Insights, helping them become more customer-centric while maximizing ROI from marketing budgets.
“We used to have to do loops and hoops to get that data, transform it into a spreadsheet, set up some systems, and it would take a couple of days to dive into the data. Diving into data is now on the fly. It’s very nearly in real time.”
Ilango Kasiraja, Senior Data Architect, Edible Arrangements
The future of Edible and Azure
Kasiraja says, “We started having a regular weekly cadence meeting with Microsoft, and they were very helpful in resolving all of our issues and getting adapted with the implementation. It has been a very great relationship.” One that Edible looks forward to continuing.
And Edible has plans for its Azure-based data warehouse and analytics platform. It wants to use Azure Synapse Link for Cosmos DB to enable real-time access to Azure Cosmos DB without impacting the transaction system. Edible wants to use Power BI to create a centralized inventory management system, and with Azure Synapse Studio, it hopes to consolidate reporting, ETL workloads, and third-party cluster-computer framework workloads into one place. Using other Azure tools, it plans to start analyzing and reporting on call center data, transaction data, and user click histories, too.
Padmanaban says, “There are endless possibilities, and the sky is the limit for all the insights that we could derive from this platform.”
Beginning September 1, 2023, Dynamics 365 Marketing and Dynamics 365 Customer Insights will merge into a single offering under Dynamics 365 Customer Insights. Learn more about the evolution of Dynamics 365 Customer Insights.
“Performance also plays a crucial factor, which the Azure tools and services brought to the table, helping us make the decision [to use an Azure-based data analytics and warehouse platform]. ”
Ilango Kasiraja, Senior Data Architect, Edible Arrangements
Follow Microsoft