Ensuring frictionless payments shoppers can bank on
Simpl speeds the delivery of ML models and data insights with Databricks
faster-to-deploy ML models
improvement in the analytics team’s turnaround time for reporting tasks
Simpl, India’s one-tap checkout and post-purchase experience platform, is on a mission to deliver secure and hassle-free transactions through advanced machine learning (ML) and artificial intelligence (AI). Catering to around 30,000 merchants and managing over 10 million transactions monthly, Simpl demonstrates a significant influence in the fintech sector, bolstered by over 100 active workflows that drive their PayLater services and various financial operations. Previously hampered by an outdated tech stack and siloed data, which caused frequent failures and lengthy ML model training, Simpl faced substantial scalability and operational challenges. The adoption of the Databricks Data Intelligence Platform revolutionized their approach, enabling a transition to real-time, scalable solutions that cut ML training from weeks to a single day and improved analytics turnaround by 35%. This transformation has not only accelerated time to market by 90% but also democratized data access, allowing a broader range of personnel to engage directly with business-critical data.
Complicated deployments and data silos stall checkout experience enhancements
Simpl has cemented their role as India’s premier checkout and post-purchase experience platform. Focused on enhancing user trust through seamless transactions, Simpl harnesses the power of machine learning and AI. Vikas Goyal, Principal Engineer leading the Data Platform team at Simpl, elaborated on their mission: “We are dedicated to offering a quick, secure and hassle-free checkout experience while building consumer trust post-purchase.”
Operating exclusively in India, Simpl collaborates with approximately 30,000 merchants and manages over 10 million monthly user transactions, showcasing their substantial market footprint. Simpl’s array of 100+ active workflows and 10K+ daily SQL queries demonstrates their robust focus on business analytics, reporting, data engineering operations and ML model training.
Their business use cases address a wide range of financial services needs. “Each workflow is pivotal in underpinning our PayLater services,” Goyal explained. “Our use cases include underwriting new users, fraud detection, user retention, churn prevention, marketing campaigns for offers, delinquency analysis, financial reporting and other critical KPI tracking.”
Simpl faced significant scalability challenges with their technology stack. As Goyal pointed out, “Our former tech stack was not just outdated but also incapable of handling our growing amount of data. We experienced frequent failures, with no common platform for access.”
The siloed data and lack of a unified platform severely hampered the accessibility and utility of business intelligence. Simpl’s teams were caught in cycles of cumbersome deployments. “Some of our ML models were taking several weeks to train. We needed a scalable solution that would reduce this turnaround time to meet the real-time demand of our analytical and reporting needs,” Goyal said. “We also wanted the ability to create dashboards and alerts on the fly to more easily query data.”
Enter the Databricks Data Intelligence Platform.
Eliminating complex deployment cycles with Databricks SQL and Databricks Workflows
Databricks has consolidated their vast datasets, transforming Simpl’s ability to harness data analytics and AI effectively. Goyal’s Data Platform team relies on Databricks SQL and Databricks Workflows to leverage the following benefits. “Databricks has significantly sped up the execution of our analytical and reporting needs, facilitated quick dashboarding for executives and expedited our ML model training processes to cater to ever-evolving data drift. We can set up alerts at various time cadences, ensuring constant monitoring of our critical business KPIs.”
With Databricks SQL, Simpl has democratized data access across the organization, enabling different teams to leverage the platform’s capabilities extensively. “Databricks SQL serves as a shared mechanism to query data, allowing quick creation of visualizations shared across various stakeholders. People with only SQL knowledge can now independently create and schedule their workloads, reducing our dependency on the DevOps team.” This component has simplified data querying and analysis, making it accessible to a broader range of users within Simpl.
Databricks Workflows has streamlined Simpl’s extract, transform, load (ETL), ML model training and reporting requirements. According to Goyal, “The ability to schedule notebooks from the workspace or Git, with no complex deployment cycles, allows people with less coding experience to write and deploy their code quickly.”
This has also enhanced the monitoring and observability of these processes, providing configurable cluster configurations depending on the workload. Goyal added, “Cluster management is uber quick and can be done in a few clicks. This comes in handy for sudden heavy workloads. The autoscaling capability helps us to reduce overall costs, keeping infrastructure scalable.”
This empowerment has led to a tangible increase in productivity and a reduction in time to market for new features and enhancements.
Data democratization accelerates time to market of ML models by 90%
Databricks has delivered measurable improvements across various departments: enhanced productivity, faster time to market and increased scale and performance. According to Goyal, “Productivity has improved across all teams. For the analytics team specifically, their turnaround time for reporting tasks has improved by 35%. And our ML models that previously took several weeks to train are now ready within one day, which marks a huge success for our data science team.”
These enhancements have also translated into tangible business outcomes, with features, enhancements and ML models shipping into production much faster — some at a rate that has reduced time to market by 90%. Now that people with varying levels of tech knowledge across teams can configure different-size clusters depending on their workload, scalability is no longer a challenge. This has allowed Simpl to be more agile in their response to market demands and customer needs.
The integration of AI/BI Dashboards has further democratized data access at Simpl, empowering a broader range of users — from C-suite executives to engineers — to analyze business trends and numbers independently. Goyal noted, “We’re better addressing use cases like growth, retention and delinquency control. Business people who previously depended on the analytics team to fetch data for them can now query data themselves, enhancing decision-making speed and accuracy.”
The successful deployment of Databricks at Simpl has not only solved their immediate challenges but also set the stage for future innovations. With Databricks, Simpl has transformed their data infrastructure into a dynamic, scalable system that drives efficiency and growth. The forward-looking strategy emphasizes continued investment in advanced technologies to maintain competitive advantage and cater to the evolving needs of the fintech market.
