Fi

Fi Uses Amazon Pinpoint to Improve Segmentation and Targeting, Sending over 12 Million Monthly Messages to Users

2022

India’s neobanking industry has grown steadily since 2019, with total funding to the sector valued at over $163 million as of Q1 2021. Neobanks are pure online alternatives to traditional banks and are typically run by FinTech firms providing financial apps for their customers. Similar to its global counterparts, neobanks in India targets specific customer segments—often individuals who grew up with digital technology or during the information age, also known as digital natives—and build product portfolios to suit their needs.
 
Fi is a neobank whose money management app targets digital natives. Within 3 minutes, users can open a savings account online, which includes cutting-edge smart features like the Ask.fi virtual assistant that specifically caters to this tech-savvy market segment. Fi currently serves more than 1 million customers and aims to help users efficiently manage their money. “We want to help customers demystify their finances, maximize savings, and spend intelligently,” says Prasanna Ranganathan, founding member of Fi.

Standing on the Shoulders of Giants

Fi chose to launch its business on the cloud in 2019 for fast time-to-value. It uses Amazon Web Services (AWS) for core compute and storage infrastructure, in addition to managed services such as Amazon Guard Duty to enhance security and AWS Config for data governance.
 
Explains Ranganathan, “With AWS, we’re standing on the shoulders of giants. All the abstractions and building blocks we need to grow our business are there. It would’ve taken us years to get anywhere near the level of sophistication and reliability we’ve come to count on in the solutions we’ve built on AWS.”
Fi
kr_quotemark

With AWS, we’re standing on the shoulders of giants. All the abstractions and building blocks we need to grow our business are there.”

Prasanna Ranganathan
Founding Member, Fi

Accelerating Innovation and Time-to-Market

When building new features, Fi engages early with its AWS account team and solutions architects for input on the proposed use case and technology suggestions. If AWS has a service addressing a target use case, Fi tries it first rather than attempting to build something in house, accelerating innovation in a fast-moving, competitive industry. “We leverage native AWS services when possible, especially machine-learning offerings. This reduces our time-to-market and iteration time,” Ranganathan says.
 
Fi uses Amazon Rekognition to validate new customers’ identity documents during the onboarding/know-your-customer (KYC) process. Engineers have also built machine-learning (ML) models in Amazon SageMaker to support other KYC functions. As part of the ID verification process, for example, new Fi users must record a video wherein they read out a random series of numbers. Fi’s ML models assess these videos alongside supporting documentation to verify identity.

Ensuring Security of Digital User Engagement Efforts

More recently, Fi has adopted Amazon Pinpoint for digital user engagement. “When evaluating engagement tools, Amazon Pinpoint was a natural choice from a data security perspective,” relates Ranganathan. “Given that our primary technology stack is on AWS, it made sense to choose a tool from AWS for maximum security, by avoiding the need to transmit sensitive financial data through another vendor or channel,” he says.
 
Members of Fi’s marketing team rely on the tool to augment their segmentation efforts and improve targeting. They’ve defined customer segments in its data lake on AWS, built using Amazon EMR, Amazon Simple Storage Service (Amazon S3), and Apache Spark. These segments are uploaded into Amazon Pinpoint, and the marketing team then simply selects the target audience for any message or campaign.

Enhancing Visibility into User Behavior, Campaign Impact

Using Amazon Pinpoint’s integrated analytics functions, Fi can then design and analyze communication at distinct points in the customer journey. For example, they can identify users who downloaded the Fi app during the past week and inform them via SMS or email about potential features of interest. Or for long-term users with low balances, they can send offers encouraging them to replenish their accounts or transact more.
 
“We’ve seen a big improvement in segmentation and targeting, and we now have a view to what’s working and what’s not,” shares Niyati Maun, product marketing at Fi. Once communications go out and campaigns are complete, message delivery data is fed back into the data lake platform via Amazon Kinesis, enriching Fi’s data set for subsequent campaigns.

Choosing the Right Channel at the Right Time

With Amazon Pinpoint, Fi has a centralized mechanism to send messages across multiple digital channels, including SMS, email, and push notifications. The marketing team can review cross-channel conversion rates, adjust communications strategies accordingly, and ascertain which time of day is most effective to send certain messages. “Amazon Pinpoint has really helped us ensure we reach the maximum number of target users with each communication, using the right channel at the right point in time,” Maun says.

The team has also streamlined their ability to run and monitor simultaneous communication campaigns. Previously, they were utilizing a semi-manual approach with CSV file uploads to set up and run each campaign, which Maun admits was tedious. With built-in automation in Amazon Pinpoint, however, they can set up five or six concurrent campaigns scheduled to run daily without manual intervention. In an average 30-day period, Fi sends 11–12 million messages to its customers.

The team also has greater visibility into which, and how many, daily communications a user receives. “We don’t want to spam users and send more than one communication per day, and Amazon Pinpoint prevents this,” says Maun. “We can prioritize campaigns within Amazon Pinpoint and rely on the service to handle intelligence around message delivery.”

Preparing to Scale 3–5 X

Currently, Fi is experimenting with A/B testing its ML models to determine the types of messages that resonate best with distinct customer segments. It’s also exploring the creation of new models in Amazon SageMaker to refine segmentation. “The freedom that AWS gives us to focus on building our product and delivering value to the customers is hard to overstate. It’s like there’s an invisible expert next to you who’s already done the hard work, so we can take what they’ve built and extend it,” Ranganathan elaborates.

Fi expects to scale its business three to five times over the next 18 months and is confident its AWS infrastructure will continue to support the business at any level. “With AWS, we have a solid base that works at any scale,” says Ranganathan. 

Learn More

To learn more, visit aws.amazon.com/products/customer-engagement.

About Fi

Fi is a money management app for the working professionals of today. Fi's core product is a savings account that can be opened entirely digitally. The account comes with a debit card and an investment option that allows automated investing.

Benefits

  • Reduces time-to-value and iteration time with managed services
  • Improves audience segmentation and target marketing
  • Successfully sends 11–12 million messages to users each month
  • Runs 5–6 concurrent campaigns without manual intervention
  • Sends and analyzes messages across SMS, push, and email
  • Verifies identity of new users with ML-assisted tools
  • Protects financial data by limiting use of external technology


AWS Services Used

Amazon Pinpoint

mazon Pinpoint is a flexible and scalable outbound and inbound marketing communications service.

Learn more »

Amazon SageMaker

Build, train, and deploy machine learning (ML) models for any use case with fully managed infrastructure, tools, and workflows.

Learn more »

Amazon EMR

Amazon EMR is a cloud big data platform for running large-scale distributed data processing jobs, interactive SQL queries, and machine learning (ML) applications using open-source analytics frameworks such as Apache Spark, Apache Hive, and Presto.

Learn more »

Amazon Rekognition

Amazon Rekognition offers pre-trained and customizable computer vision (CV) capabilities to extract information and insights from your images and videos.

Learn more »

Get Started

Companies of all sizes across all industries are transforming their businesses every day using AWS. Contact our experts and start your own AWS Cloud journey today.

Contact Sales