How the Infosys Customer Intelligence Platform Delivers a World-Class Customer Experience
By Arav Narasimhamurthy, Sr. Principal – Infosys
By Mithun Das, Sr. Technology Manager – Infosys
By Jignesh Desai, WW Migration SA – AWS
By Navtanay Sinha, Sr. Product Manager, Amazon Neptune – AWS
Ease of use and hyper personalization of customer-facing applications are essential in today’s competitive landscape. With digital transformation as the driver for enterprises, customer experience is the key differentiator for revenue, satisfaction, and loyalty.
Today’s customers demand service tailored to their unique needs. To meet these expectations, enterprises must continuously learn, evolve, and develop mechanisms to drive usability and improve the omnichannel experience.
For example, in the case of a car accident, customers want to be able to file a claim with their auto insurance, share details, and upload images from the accident site—all via an app on their smartphone. In this scenario, the customer expectation is that the claim should be approved within a few hours. They also want to receive push notifications for nearby auto repair shops, including reviews and the ability to easily schedule an appointment.
However, most auto insurance providers today—even with cutting-edge web and apps for claims and registration—don’t provide fully qualified automated recommendations for nearby facilities.
In this post, you’ll learn about the Infosys Customer Intelligence Platform (CIP), a solution built using Amazon Neptune and Amazon Web Services (AWS) to accelerate data ingestion, data processing, data modeling, and data analytics.
The Infosys Customer Intelligence Platform leverages data (events, interactions, external, telemetric data, and IOT), builds knowledge graph (customer identity, products, financial solution needs, risk, and preferences), and applies machine learning (ML) algorithms to understand data from a multi-dimensional perspective.
It also provides a unified experience, enabling users with deep insights, an omnichannel experience, hyper-personalization, and a 360-degree view of the customer and product. The solution primarily is comprised of three building blocks, including:
- Modernization: CIP takes an in-depth look at legacy customer journeys with manual inputs and automates it with use of AWS cloud-native services. Once the transactional data is ingested, it streams data to Amazon Kinesis for real-time ingestion and insights. Once data is collected, it’s streamed for further processing using AWS Glue. This activity is used for unified semantics. AWS Glue processes the data and creates a map with meta-data collected from all the manual inputs and stores it in Amazon Redshift for analytics and Amazon Simple Storage Service (Amazon S3) for long term storage.
- Intelligence: At the heart of the CIP is a digital brain which constantly builds, trains, and deploys ML models using Amazon SageMaker to drive intuitive decisions. CIP also uses Amazon Neptune knowledge graph, a database and compute, to map the many-to-many relationships between the entity and other entities, such as the policy holder and claims adjustor, among others. This also helps with the analysis of data, images, documents, and unstructured information, as well as recommending decisions.
- Experience orchestration: Each manual input from customers in their flow of an application is recorded using functional set of meta-data. This capability is useful in anticipating and prioritizing the most needed feature for customers. Personalized campaigns and nudges helps with all points of interactions with customers. CIP also automates the business process using adaptive supply chain, promotion, and channel monetization.
The platform is built working backwards from the customer experience and using the following tenets to drive business outcomes:
- Smart platform with the ability to capture customer signals, events, interactions, and phygital data in real-time events, interactions, external, telemetric data, and IoT.
- Build knowledge graphs for customer identity, products, risk, and preferences.
- Anticipate the needs of applicants and businesses including real-time product recommendation. The platform can understand an individuals’ needs from existing data to sense, respond, and automate routine decisions.
- Offers a unified experience providing users with hyper-contextual insights, omnichannel experience, hyper-personalization, and a 360-degree view of the customer and product.
- The platform can also reimagine and bring zero latency in key business processes, such as completing the quote process in a few steps and persona-based application handling to reduce the completion duration of complex cases.
Architecture and Inner Workings
As shown in Figure 1, transactional and real-time data coming from various events are ingested using Amazon Kinesis. AWS Glue and Amazon EMR are used to process the data, with initial processing stored in Amazon S3 and Amazon Redshift for deeper analytics.
Amazon Athena is used to do ad-hoc analysis on raw data stored in Amazon S3. The platform uses Amazon SageMaker ML models and Amazon Fraud Detector along with Amazon Personalize and Amazon Rekognition for auto damage detection, auto policy recommendation, and fraud detection.
The platform ingests data from customer claim history, vehicle information, geo-events, and third-party data to build a knowledge graph using Amazon Neptune. This helps capture key events and enables intelligent decision making; which in turn is used to improve the customer experience for subsequent claims and product recommendations.
Amazon QuickSight provides native connectivity to Amazon Redshift and Amazon S3 which is used in visualization of churn, processing times, and next steps.
Figure 1 – Infosys Customer Intelligence Platform architecture.
The CIP Platform integrates seamlessly with legacy customer facing applications and provides the following benefits:
- Enables enterprises to jumpstart their analytics journey and provides an agile platform for data driven decision making across business functions.
- Improvement in agent productivity with seamless, just-in-time insights.
- Increased marketing campaign effectiveness with increase in cross-sell and upsell with quick conversion.
- Customer experience score improvement with next best action helps in reducing delay to resolution by 30%.
To accelerate the journey to Amazon Neptune, Infosys built the following key elements:
- Automated data pipeline to ingest data from existing databases and graph database to Amazon Neptune.
- Utility to export Neptune graphs from cloud to local for offline exploration, augment with more data from other sources as applicable.
- Easy graph visualizer in Neptune workbench for viewing the exported graphs—nodes, edges, properties, and query builder interface.
- Load data at one-shot to Neptune with real-time status of loading, completion, and graph node count.
- Open Jupyter notebook right from the scene to view and query the graph loaded in Neptune.
- Export the Neptune graph to Amazon S3 bucket(s) if applicable.
Today’s customers demand seamless, personalized experiences—and they expect enterprises to have the insights to make timely and informed decisions based on their unique needs.
In this post, we explore how the Infosys Customer Intelligence Platform is designed to improve the customer experience. Adoption of AWS cloud-native services helps enterprises correlate scattered siloed data, contextualize on manual customer inputs across physical and digital systems, create mind-map around events, and respond quickly to hyper-personalized decisions.
The platform continuously evolves using existing data and new stream of data with learning to drive intuitive decisions leveraging knowledge graph, AI, and intelligent ops.
For more details and implementation of the Infosys Customer Intelligence Platform, contact the Infosys team.
Infosys – AWS Partner Spotlight
Infosys is an AWS Premier Tier Services Partner that helps enterprises transform through strategic consulting, operational leadership, and co-creation of solutions in mobility, sustainability, big data, and cloud computing.