Aavas Financiers Limited was originally incorporated as a private limited company on February 23, 2011, as "Au Housing Finance Private Limited" at Jaipur, Rajasthan. As a housing loan provider, it serves largely underserved and untapped markets. The company's housing finance business has three primary products: Finance for Purchase, Construction, and Home Extension. It provides housing finance to small families in rural areas, towns, the city's peripheral area and other semi-urban areas so that they can achieve their dream of owning a home. It operates in 13 states and has over 300 branches. The technology teams are based in Jaipur, Rajasthan.

Manish from Aavas was interviewed for the case study. As AVP, Data Engineering, he has been with Aavas for more than 7 months. He has over 13 years of experience in data engineering, including data warehouses, data lakes, and cloud infrastructures, such as Azure and AWS. He has also led many projects involving data pipelines, cloud migrations, and management information systems (MIS). He has assisted organizations in making informed, responsible decisions based on data.

The data engineering team that Manish heads is based in Jaipur and has 7 members. The team designs information processes for many types of data tasks, such as the acquisition of data, data transformation, and data modelling. It also delivers data in a format useful to the data science and analytics team. The analytics team then builds reports and dashboards that are used across the organization. At Aavas branches across India, daily reports and dashboards covering important metrics, and are integral to the operations.

The initial set of problems was multifaceted. To make data-driven decisions, most teams required data. The teams would query the production databases in the absence of data warehouse infrastructure. It was inefficient to query the production database directly for reporting. Therefore, the data team wanted to build a data warehouse to handle all these reporting tasks.

The next challenge was strategically designing data pipelines incorporating the data warehouse. The data needed to be ingested from over 10+ data sources. Some of these data sources were core to the business, like the Loan Management System (LMS), Customer Relationship Management (CRM), etc. After the data was ingested, an interface had to be created to allow teams across the company to generate integrated reports and insights from the ingested data. Users in the organization were not all knowledgeable about how to query databases to find the information they needed. Hence, it was important to find a solution that could be used by all stakeholders. Thus, the data engineering team at Aavas were on the lookout for a solution that would meet all their needs.

The data engineering team contacted the Sprinkle team. Together, they talked about challenges the data team faced and the problems that needed to be solved. Sprinkle's solution team demonstrated a proof of concept (POC) to the executives by pitching Sprinkle's No-Code Integration and Analytics Solution after understanding the use cases. To ingest data from multiple systems, Sprinkle's ready-to-use connectors with advanced features such as real-time data replication and schema mapping were used. The transformation and analytics layers were then used to get the data in the required format and build reports and dashboards.

Data from systems like LMS, CRM and other in house systems that use databases like MySql and MongoDB are being ingested into the on-premise Hive data warehouse. Data from the warehouse is used to develop various models for teams such as the credit and loan department. Reports are being created and shared regularly with the teams and the executives using these models.

Reports and dashboards created using Sprinkle are being consumed by over 30+ Aavas branches across India. These reports track various metrics that are important for the day to day functioning of the branches. These are also used by the senior executives of the company to track the performance of the company. And thus, take various financial and administrative decisions regarding the product and the operations.

Manish says, “It’s very quick and easy to set up data and build models, reports and dashboards on top of it. We can now utilize the saved time to innovate on other problems”.

Since the data gets automatically ingested into the data warehouse, it is very easy to build new models, without worrying about the data. Thus, improving the productivity of the data team. Ad-hoc data requests have decreased as users from business teams are also able to get the relevant data using Sprinkle. The automatic ingestion of new data and the auto-refresh feature of reports and dashboards make updating them very easy, reducing the need for manual work.

Manish says, “Sprinkle is extremely easy to use and has a high-quality look and feel. The clarity in the purpose of various data modules makes it easy to create solutions for our numerous use cases”.

Presently, the data engineering team at Aavas is working to migrate from on-premise data warehouses to cloud warehousing systems. Sprinkle is helping them understand how to easily transition to the cloud data warehouse.