AWS for Industries

How AWS is supporting Buy Now Pay Later (BNPL)

The concept of purchasing an individual item in multiple installments is not new. Large department stores have offered this for decades in what is called layaway. Layaway is essentially “a purchasing agreement by which a retailer agrees to hold merchandise secured by a deposit until the price is paid in full by the customer” [1]. The notion of Buy Now Pay Later (BNPL) is similar, the only difference is that the merchant releases merchandise at first deposit, with intention to receive multiple installments paid thereafter. Despite not being novel, BNPL is accelerating; according to a recent survey of U.S. consumers, 83% of consumers use BNPL to make large ticket service purchases and note that it is a practical alternative to using personal loans and credit cards. Furthermore, 61% of consumers are using BNPL to help manage their spending and budget on purchases by fragmenting payments over time [2].

See full research on The Next BNPL Horizon Will Expand Access to 83% Who Want to Make Big Ticket Purchases.

What exactly are the primary use cases for BNPL?

There are two key use cases where BNPL resides:


The most popular use case out of the two is ecommerce. Through online checkout BNPL functions as an alternative payment option to the traditional payment selections. Consumers simply choose BNPL as a payment plan and select a repayment option. This occurs after a rapid soft credit check.

 Point of Sale (POS)

Another channel is through brick-and-mortar retailers using POS systems that can offer a credit at instore checkout, typically through a QR code or a payment link. (See how AWS supports QR here).

How can Cloud help support BNPL?

Financial Services institutions can leverage AWS to create BNPL journeys in a few ways, primarily in the following four areas:

  • API and core modernization: Building standardized APIs across services that enable rapid and seamless integration with third parties and partners.
  • Extending credit: Managing, analyzing, and applying machine learning tools to large and real-time data sets for more effective lending decisions.
  • Integration of 3rd party data sets: Utilizing third party data to augment internal data and enhance customer engagement and credit.
  • Customer Service: Streamlining customer service through self-service and assisted channels, and leveraging data and usage patterns to recommend new products and services.

In addition, following is a detailed toolbox that details AWS specific services for BNPL.

 The AWS Toolbox for BNPL

  • Amazon API Gateway is a fully managed service that makes it easy for developers to create, publish, maintain, monitor, and secure REST APIs at any scale.
  • AWS Fargate is a serverless compute engine for containers that works with both Amazon Elastic Container Service (ECS) and Amazon Elastic Kubernetes Service (EKS).
  • AWS Lambda lets you run code without provisioning or managing servers. You pay only for the compute time you consume. With Lambda, you can run code for virtually any type of application or backend service.
  • Amazon Cognito is a user management service with rich support for user authentication and authorization. You can manage those users within Amazon Cognito or from other federated IdPs.
  • Amazon DynamoDB is a key-value and document database that delivers single-digit millisecond performance at any scale.
  • Amazon Pinpoint makes it easy to run targeted campaigns and drive customer communications across different channels: email, SMS, push notifications, or custom channels.
  • Amazon Sagemaker is a fully managed service that provides every developer and data scientist with the ability to build, train, and deploy machine learning (ML) models quickly
  • AWS Glue is a serverless data integration service that makes it easy to discover, prepare, and combine data for analytics, machine learning, and plication development.

 Reference architecture

The following BNPL solution architecture diagram demonstrates the flow of consumers interaction with the system.

User Service: User service gathers user attributes such as name, phone number, and email address that manage logged-in or registered user functionality.

  1. Customer selects an item(s) for purchase and prepares to check out. At the check-out process, they are then presented with available BNPL options. Customers proceed by selecting the plan which best fits their needs.
  2. The request is routed to a separate page which takes the user through the BNPL flow. This web page can be built using AWS Amplify which is tightly integrated with Amazon Cognito and has pre-built UI workflow components for new customers to sign-up and existing customers to sign-in.
  3. Amazon Cognito integrates with your client code, aids in authentication and authorization workflows.
  4. Amazon Cognito User Pools are integrated with Amazon Pinpoint that provides multi-factor authentication notifications, analytics and targeted campaigns to drive user engagement in mobile apps using push notifications.
  5. New authenticated customers are verified by sending a notification to their devices using Amazon Pinpoint.
  6. Authenticated clients make API calls to AppSync using valid JWT tokens generated by Cognito.
  7. AppSync uses Resolvers to make direct calls to different microservices. HTTP Resolvers connect to REST endpoints of the User service. A Lambda Resolver directs calls to the Credit Service in a VPC.
  8. The communication between the Resolvers and the HTTP endpoints are protected with temporary IAM credentials based on assumed IAM roles. The JWT token specific to the authenticated user is also forwarded to each microservice.
  9. The REST microservice returns the requested information in XML based on the user details in the JWT Token. The XML payload is automatically converted to JSON by built-in utilities in the AppSync resolver.

Credit Service: This is a private service in a separate VPC only accessible from an authorized request aids in credit processing.

  1. A Lambda function is invoked to access the private service hosted in a VPC via AWS PrivateLink. PrivateLink provides private connectivity between Credit Service VPC and Lambda on the private AWS network.  All services are secured in a way so that only the main AppSync API is granted access.
  2. The Credit service is hosted on AWS Fargate containers in a private VPC and payment information is persisted in DynamoDB. Customer requests are evaluated against their pseudo credit rating via a service from the Amazon SageMaker elastic inference and provided a real-time decision.
  3. The response on payment information is notified to the user and also via notification subscribed in the applicatio

Credit Scoring: This service aggregates and enriches data available from various credit sources to make more informed decisions.

  1. Data from multiple credit agencies is streamed in real time to an Amazon Kinesis data stream. AWS Lambda function pulls data off this Kinesis stream, and triggers a workflow.
  2. The Step Function workflow goes through a series of steps by first loading the raw data in S3. Next, AWS Glue performs an ETL job to transform the data. After that, using Amazon Sagemaker it trains and deploys a machine learning model on this data to make a credit decision.
  3. Decisions and values against features are audited at the time of decision and stored for operations teams to provide customer feedback and compliance reporting.Conclusion


Financial services institutions are at the early stages of “Buy Now Pay Later.” Various firms are now finding better, faster ways to serve consumers with and without a credit history to help increase revenue opportunities. Leveraging the cloud, customers are able to offer services like Machine Learning to help reshape the accuracy and speed of underwriting, as well as modeling consumer credit risk. Additionally, they can construct standardized APIs across a myriad of services that enable rapid and seamless integration with third parties and partners. Customers like Affirm have used AWS to enable BNPL to manage, analyze, and apply machine learning tools to large and real-time data sets for more efficient lending decisions.

For more information about how to work with AWS and to understand how we are supporting payment customers around the world to address niche payment methods, please contact your AWS Account Manager or visit AWS Financial Services – Payments.




Sekai Ndemanga

Sekai Ndemanga

Sekai Ndemanga is a Payment Business Development Manager at AWS with a specialty in Fintech payment customers. She helps financial services customers navigate new payments regulations, emerging payments technologies, mobile payments, and traditional card and banking services.

Sudhir Kalidindi

Sudhir Kalidindi

Sudhir Kalidindi is an AWS Principal Solutions Architect in Financial Services with 22+ years of experience in software architecture and the development of solutions involving business and critical workloads. He helps payments customers to innovate on the AWS Cloud by providing solutions using AWS products and services.