Who is the customer?
A shopper buys running shoes on your website using Gmail. Two weeks later, they join your loyalty program with their work email. Next month, they make an in-store purchase and provide a different phone number. Your systems now see three separate customers—but it’s the same person. Multiply this scenario across millions of transactions, and you have the retail industry’s most expensive data problem: fragmented customer identity. For retailers and consumer goods brands, this fragmentation has real business consequences—wasted ad spend on duplicate audiences, missed personalization opportunities, and an incomplete view of customer lifetime value. In this post, we explore how Amperity and Amazon Web Services (AWS) are helping retail organizations unify customer identities to drive measurable business outcomes.
Why traditional matching falls short
Traditional deterministic ID matching assumes a single email or phone number uniquely identifies a person across systems. In reality, a single shopper might use one email for online purchases, another for a loyalty program, and a different phone number in-store. They may share a household address with family members who are also customers. Traditional deterministic matching—linking records by a single identifier like email struggles with this challenge.
Retailers see the same customer as multiple people, or multiple customers as one. This leads to redundant marketing spend targeting the same person across channels, inaccurate customer lifetime value calculations, poor personalization that erodes brand trust and missed opportunities to identify high value omnichannel shoppers. As retailers invest heavily in AI-driven personalization, the quality of the underlying customer data becomes the critical foundation.
Meet Chuck Data
More formally, meet Chuck Data which is a command-line AI agent for customer data, built by Amperity. It embeds years of customer data expertise and multi-patented identity resolution algorithms. Trained on billions of datasets across 400+ enterprise brands, Chuck Data is a tool that data engineers can run directly in their existing environment. Together with AWS, Chuck Data helps retailers unify fragmented customer data into accurate, actionable profiles—without requiring data to leave the customer’s AWS account. Chuck Data securely authenticates with both Amperity’s API and your AWS account keeping data within your environment. This means organizations can add identity resolution to their existing AWS infrastructure without introducing external dependencies or moving sensitive customer data outside their environment. The collaboration brings together the following key points:
- Amperity’s Stitch technology’s patented AI-powered identity resolution, which uses machine learning to probabilistically resolve customer identities across data sources—going beyond simple rules-based matching to create a first-party identity graph.
- Amazon Bedrock for foundation model access. Through Bedrock, customers interact with data workflows using natural language and can choose from 100+ LLMs based on their specific needs—including Amazon Nova, which offers strong cost efficiency for high-volume data workflows.
- Amazon Redshift as the data warehouse layer, allowing identity resolution to run directly against existing Redshift tables—no data migration required.
- Amazon EMR for compute running within existing VPC configurations and leveraging the IAM policies and governance controls retailers have already established.
Testing Chuck Data’s impact with sample retail data
To evaluate Chuck Data’s capabilities on AWS, a sample retail dataset of approximately 1.4 million records—based on actual customer data—was deliberately corrupted with malformed entries, mislabeled columns, inconsistent formats, and overlapping customer identifiers. Chuck Data detected the issues, flagged inconsistencies, and offered automatic fixes. The test produced the following results:
Figure 1: Example Chuck Data console
Figure 1 shows the Chuck Data CLI `/status` output displaying the current configuration — connected to an Amazon Redshift workgroup in `eu-north-1`, using Amazon Nova Pro (`amazon.nova-pro-v1:0`) as the foundation model, with a confirmed connection to 3 databases.
43% fewer duplicate records
Figure 2: 43% fewer duplicate records
The sample dataset, shown in Figure 2, contained 139,228,651 source records, which Chuck Data resolved into 79,708,992 unique customer clusters — a 43% reduction in duplicate records
83% known customer profiles
Figure 3: 83% known customer profiles
Figure 3 shows the Record Level PII Classification Improvement matrix, which displays how customer profiles were reclassified. Rows represent pre-resolution PII classifications (unknown, anonymous, partial, known) and columns represent post-resolution classifications. The darkest cells along and above the diagonal show upward mobility — notably, 24,645,876 partial records were upgraded to known, and 90,366,711 records retained their known status. Post-resolution, 115,745,278 of 139,228,651 total records (83%) achieved “known” classification, demonstrating improvement in customer profile completeness.
In summary, Chuck Data, powered by AWS, handled the anomalies by cleaning and correcting them in roughly 10 minutes. Further unifying the data completed in just over 14 minutes.
Why it matters for retail and consumer goods organizations
The combination of Chuck Data and AWS addresses several priorities for retail leaders –
Data stays in your environment: Identity resolution runs within your own AWS account. Chuck Data runs locally and executes within your cloud infrastructure — your data, credentials, and security configuration are never accessed externally. For retailer organizations operating under strict data governance and privacy regulations, this architecture means there’s nothing new to vet beyond your existing infrastructure.
Faster time to value: Rather than months-long data migrations, retailers can run identity resolution against data already in Amazon Redshift or connect through Amazon EMR (Elastic MapReduce). Chuck Data’s natural language interface means data engineers can complete days of coding with a single prompt—getting from setup to unified customer profiles in minutes, not months.
Generative AI ready: With Amazon Bedrock integration, retailers can use natural language to interact with their customer data workflows—profiling data, standardizing PII, and running identity resolution without writing code. This lowers the barrier for data teams to leverage AI capabilities.
AWS partner overview: Amperity
Amperity is an AWS Retail Industry Partner that helps brands unify fragmented customer data to drive personalized engagement at scale. Amperity Customer Data Cloud, available on AWS Marketplace, delivers AI-powered identity resolution across retail and consumer goods. Chuck Data is available as a free, open-source CLI tool that runs natively on AWS.
So, who is the customer? Now you know.
The question “Who is your customer?” shouldn’t be hard to answer. With Chuck Data and Amazon Bedrock, it isn’t. By combining Amperity’s Stitch technology with Amazon Bedrock’s LLM capabilities, you can transform fragmented data into unified, actionable customer profiles—all while keeping sensitive data within your existing AWS infrastructure.
The retailers winning in today’s competitive landscape aren’t the ones with the most data. They’re the ones who actually know who their customers are. Stop paying to market to the same customer three times. Stop missing high value omnichannel shoppers. Stop building AI personalization on a foundation of fragmented identities. Instead, start seeing your complete customer. Start maximizing every marketing dollar. Start building personalization that works.
Ready to get started? Download Chuck Data from the Amperity website. Explore Chuck Data on GitHub for documentation, demos, and setup guides. For more information about Amazon Bedrock, see the Amazon Bedrock documentation. To learn more about identity resolution best practices, see the Amperity documentation on Stitch technology.
Download Chuck today and experience customer data unification the secure way, the accurate way, the smart way.
