Data-Driven Customer Management Across Data Silos, Powered by PwC’s Customer Insights Engine on AWS

By Kiran Kumar Ballari and Rama Lankalapalli – AWS
By Mo Bashir, Ankur Goyal, and Karthikeyan Chokappa – PwC Australia

PwC

Customer insights are essential for businesses that want to provide a different and unmatched experience to their customers and stay competitive. Typically, organizations store data from different departments and lines of businesses in various systems, making it difficult to get a comprehensive integrated view of the customer across the value chain. Not all organizations have the expertise to collect, analyze, and interpret customer data.

To fully engage with customers and produce meaningful experiences, businesses are aiming to achieve a single unified 360-degree view of the customer. Organizations are looking for simpler ways to leverage untapped data in disparate systems and use modern technologies to generate a integrated customer view enabling personalized analytics and making informed decisions.

This means provisioning unified, real-time, secure and unique customer view with role-based access to customer data, so that businesses can focus on enhancing their customers’ experience rather than building and operating a accelerator on their own.

To address this, PwC Australia developed the Customer Insights Engine (CIE) on Amazon Web Services (AWS) that provides end-to-end business outcomes with a readily-available analytics engine powered by machine learning (ML) and integrated with out-of-the-box analysis templates, ingestion, and transformation patterns and data pipelines.

The CIE accelerator brings key customer information and transactions from multiple source systems to generate a unified customer view, enabling personalized analytics and making informed decisions.

In this post, you’ll learn how PwC CIE accelerates the process of generating a distinctive golden record for a customer from various sources. We will delve into the architecture and showcase how CIE enriches integrated insights for a customer across different operations for an enhanced customer experience.

PwC is an AWS Premier Tier Services Partner with AWS Competencies in Data and Analytics, DevOps, Security, and many more key areas of cloud computing. PwC helps you drive innovation throughout IT and the business to compete in today’s service economy.

Accelerator Overview

PwC CIE is aimed at the creation of a golden customer record by aggregating all data about that customer from across channels that will be connected to provide a personalized customer experience, allowing a customer to move seamlessly through a tailored journey.

Figure 1 – PwC CIE accelerator overview.

PwC CIE creates a master customer record, termed a Customer Golden Record (GR), by matching and merging customer data from multiple source systems. It uses a fuzzy logic-based algorithm on key customer parameters and thereby enables a single analytical view for each customer across disparate sources, both internal and external.

Many organizations lack the necessary time and resources to create, enhance, and integrate customer insights. CIE provides pre-built capabilities and automated integrated workflows to free up time and resources. The accelerator is an AWS cloud-native platform that can ingest, store, catalogue, meta tag, transform, and deliver insights at scale.

CIE offers a centralized digital environment with the following capabilities:

• Access to environments in minutes: The entire accelerator has been developed using infrastructure as code (IaC) templates, allowing client environments to be created on demand. With a simple interface, administrators can initiate deployment pipelines and manage permissions.
• Ease of use: The configuration files, coupled with full and delta load capability, provides the user with the flexibility to add, delete, or update the definition based on the organization’s evolving requirements.
• Extensible: Customers can adapt the latest AWS services as they become available or integrate with third-party data sources. The extract, transform, load (ETL) and GR pipeline is configurable and can be extended to include additional attributes and sources.
• Metadata-driven transformation pipelines: The YAML-based configuration file provides the interface to easily define the requirements.
• Fuzzy logic-based match and merge algorithm: Customer linkage model through the formation of Customer Golden Record that links customer data across all touch points.
• Graph data builder: Automatically creates the graph edges and nodes per the requirements defined through the graph configuration file.
• Environment monitoring, governance, and observability: Implemented through AWS cloud-native services with ability to integrate with customer enterprise systems.
• Near real-time updates: The accelerator pipeline is an event-driven pipeline that updates the incremental data into the analytics data store.

Accelerator Architecture

The PwC CIE accelerator is designed and developed using AWS-native services. It enables a true Customer 360 model, which links a customer across all touchpoints and builds the Customer Golden Record, in a manner that is easily customizable and updates near-real time. The accelerator architecture uses core AWS services to create GRs across various sources providing a real-time insights and recommendations.

• • Multiple line of business (LOB) data are connected, along with external data sources where the predefined data columns are stored in a common datastore backed by Amazon Simple Storage Service (Amazon S3) for driving analytics, personalization, and marketing.
  • Data from various sources is ingested in near real-time events, driven by Amazon EventBridge and transformed using metadata-driven transformation pipelines using AWS Glue.
  • Amazon Macie is used to discover and protect any sensitive data in S3. The data is transformed, cleansed, and enriched, and a golden copy (GC) is secured in S3.
  • AWS Step Functions orchestrate the workflow by persisting the GC in Amazon Neptune, a graph database displaying a unique customer record across various sources providing a real-time insights and recommendations.

The diagram below depicts the high-level architecture of the CIE accelerator and workflow on AWS:

Figure 2 – PwC CIE accelerator architecture and workflow overview.

1. When a change is detected in the input and config data store by Amazon EventBridge, it triggers AWS Lambda to update the S3 cache bucket.
2. Lambda updates S3 bucket cache with a list of table names and config file changes based on the information passed by EventBridge.
3. Amazon EventBridge rule that runs on a schedule triggers Lambda to periodically check S3 bucket cache. If there are items found in the cache, it will pass it to AWS Step Functions and then clear the cache.
4. Amazon Macie is used to run a set of pre-defined data quality checks to generate data quality report, which will be notified through SNS and will be used as feedback to fix issues at source.
5. AWS Glue Crawler scans the input data and creates AWS Glue Data Catalog, which will be used as the Hive metastore. The ETL pipeline uses this to process the input data and build the CIE dataset
6. PySpark-based Glue job runs metadata-driven customizable data transformations (selecting attributes of interest and deriving additional attributes).
7. Glue job runs the match and merge algorithm for building the Customer Golden Record based on the GR configuration that provides attributes and its importance for scoring match and merge outcome.
8. Glue job runs creating schema during full load and adding/updating key-value pair for full and delta load into Amazon DynamoDB.
9. Glue job creates files for each node with its associated properties and for each edge types. The files are loaded into Amazon Neptune as a full load always.

Unified Customer View: Golden Record Overview

PwC CIE can be used for all industries that have a requirement to keep and assess customer analytics. The following illustration dives deep on the standard retail industry use case.

Figure 3 – Unified Customer View – Golden Record Creation

As an example, take a retail organization (as shown in Figure 3) that has multiple LOBs such as physical store, online store, and insurance business. Each source system or system of record is typically having its own customer master data. The CIE-driven Golden Record establishment process links the customer information or the customer master data from different source system.

CIE will then provide a trusted, single view of a customer across dimensions by connecting several facets that define a customer (name, address, contact information, gender, preferences and interactions with various business). It enables to create one source of truth, or one customer master data across the various line of business in an organization

Key Benefits of PwC CIE

• Provides a single unified view of the customer with multiple attributes derived by artificial intelligence (AI) and machine learning techniques.
• Enables personalized analytics, customized customer experience to make informed decisions.
• Flexibility is at the core of the engine, and will connect easily to all standard business systems.
• Near real-time customer event store to enable self-service reporting and analytics.

Conclusion

To help organizations learn more about their customers, PwC Australia developed the Customer Insights Engine (CIE) accelerator on AWS.

In this post, we showed how CIE can transform customer data silos into fuzzy matching data by creating a golden copy record. By matching and merging customer data, it creates a master customer record linkage model through the formation of Customer Golden Record that links customer data across all touch points.

To learn more and begin using CIE on AWS for your customer data analytics, sign up for PwC Customer Insights Engine on AWS Marketplace.

.

PwC – AWS Partner Spotlight

PwC is an AWS Premier Tier Services Partner that helps you drive innovation throughout IT and the business to compete in today’s service economy.

Contact PwC | Partner Overview | AWS Marketplace | Case Studies