Reserveout is a restaurant reservation platform with web and mobile apps that enable users to browse, discover, and instantly book a table at a restaurant. Reserveout also is a tool for restaurants and restaurateurs. Restaurants are supplied with a digital panel for reservation and table management, client relationship management (CRM), and online marketing modules. Currently, Reserveout is the largest reservation platform in the Middle East.
Reserveout’s regional growth has led to large traffic volumes, reaching more than 340k restaurant visits per month. Data is gathered from different sources (RO terminal, mobile application, mobile site, website, Google Maps booking, and more), and stored in Amazon RDS database. The data needed proper mapping to get readable visualizations. Reserveout needs to understand and explore their data, allowing them to make better data-driven decisions. Moreover, the analytical dashboard needed for business intelligence, Reserveout needs to store the data in a place where it could be used for machine learning projects in the future.
Amazon Relational Database Service (RDS) operates the databases for Reserveout. RDS is running in the aws-reserveout production account, and for analytics we created aws-reserveout-analytics account. Amazon RDS Read Replica was provided in the production account to prevent additional traffic on the RDS instance, therefore enhancing performance. In order to access the RDS Read Replica in the aws-reserveout production account, we created a role in the aws-reserveout-analytics account for the AWS Lake Formation with delegated access across the production account.
Using AWS Lake Formation, we defined the RDS Read Replica as the data source and collected the raw data into S3 data lake. Then we used AWS Glue to perform ETL (extract, transform, and load) job. AWS Glue made it easy to prepare the data for analytics and save it as processed data. Athena enabled us to perform interactive queries for the processed data in S3. QuickSight allowed the team of data analysts to build an interactive dashboard assisting business analytics at Reserveout to draw insights from their data and make data-driven decisions.
The automation of our solution is based on two main components. The first component is AWS lake formation blue print which allows us to build a work flow for data ingestion and customize a trigger for the mentioned purpose (e.g. customizing weekly trigger for data update). The second component is Glue ETL Workflow which can be modified to run once the data is updated and a flag is checked.
The dynamic dashboards have helped the board of directors
better understand and analyze the data thereby increasing the
efficiency of their decisions. The dashboard enabled them to
be more data-driven and guided strategic business decisions
that align with their goals, objectives, and initiatives. The
proper data visualizations helped managers with architecting
and shaping their operations to be more aligned with time
framework of the company.
Additionally, after implementation of the artificial
intelligence and advanced analytics solution, ReserveOut has
noticed nearly 35% reduction in the time that is spent on
operations management. This allowed them to free up valuable
resources to concentrate on further streamlining their core
operations. With the time saved, they were able to realize a
return on their investment within the first 6 months!