Profile unification using CDP templates and customization

Overview

Profile Unification (Golden Record creation) in Redpoint Customer Data Platform (CDP) delivers a single, trusted, and actionable customer view by consolidating, standardizing, and resolving identities across multiple data sources. Redpoint’s implementation leverages pre-built CDP templates that accelerate deployment and ensure consistent data hygiene and governance, while still allowing for extensive customization to match each client’s data model, business rules, and downstream activation requirements.

Base CDP template architecture

Core components

The out-of-the-box Redpoint CDP template includes:

• Source input schemas: These are predefined ingestion templates designed to streamline the process of collecting and organizing various types of data. They specifically cater to customer, contact, transactional, and digital event data, ensuring that the information is accurately captured and ready for further processing.

• Standardization layer: This component implements normalization rules that standardize key elements such as names, addresses, emails, and phone numbers. By applying these rules, the system ensures that data is consistent and can be easily compared and analyzed across different datasets, which is crucial for maintaining data integrity.

• Identity Resolution engine: This sophisticated engine employs both deterministic and probabilistic matching logic to accurately identify and link customer records. It utilizes configurable rulesets that can be tailored to meet specific business needs, allowing for a more precise resolution of identities across various data sources.

• Golden Record table: The Golden Record table consolidates all relevant customer entity information, applying survivorship logic to determine which data points should be prioritized. It also includes standardized attribute mapping, ensuring that all data is uniformly represented and easily accessible.

• Operational metadata: This aspect of the system includes a match audit trail that tracks the history of identity resolutions, source lineage that outlines where data originated, and quality scoring that assesses the reliability and accuracy of the data. This metadata is vital for maintaining transparency and trust in the data management practices.

Default data model

The default template implements a canonical Party → Contact → Transaction → Event model:

• Party: A unique individual or household entity that serves as the primary record within the Golden Record table. This entity is crucial as it represents the core identity of the customer, allowing for a comprehensive understanding of their interactions and preferences.

• Contact: This refers to the various channel-level details associated with the party, including but not limited to email addresses, mobile phone numbers, and physical addresses. These contact details are essential for effective communication and engagement with the customer, ensuring that outreach efforts are personalized and relevant.

• Transaction/Event: This encompasses the behavioral inputs linked through identifiers such as CustomerKey or PartyID. These transactions or events reflect the actions taken by the party, providing valuable insights into their behavior and preferences. Analyzing this data helps in tailoring services and marketing strategies to better meet the needs of the customer.

• Identifiers: These are cross-system IDs that include various forms of identification such as CRM ID, Loyalty ID, Device ID, and others. Identifiers play a significant role in connecting different systems and databases, ensuring that all interactions and transactions can be accurately tracked and attributed to the correct party. This integration is vital for maintaining a comprehensive view of the customer journey across multiple platforms.

Unification workflow

Ingestion and standardization

Data is ingested through Redpoint Data Management (RPDM) workflows or via automated connectors into standardized staging tables.

Standardization rules normalize casing, remove punctuation, and apply data hygiene routines such as:

• Email lowercasing, alias correction, and domain normalization: These processes help maintain consistency and accuracy in email communications. By converting all email addresses to lowercase, we eliminate potential discrepancies caused by variations in case sensitivity. Additionally, correcting aliases ensures that all communications are directed to the correct recipients, while normalizing domains helps in managing and categorizing email addresses effectively.

• Phone number formatting: This process utilizes standard country rules to ensure that all phone numbers are presented in a uniform format. This practice not only enhances readability but also aids in the efficient processing of contact information. By adhering to the established formatting guidelines for different countries, we can minimize errors and improve the overall quality of our contact databases.

• Address cleansing: This process employs Redpoint’s address verification routines, which involves validating and correcting addresses to ensure they are accurate and deliverable. By implementing rigorous address verification, we can significantly reduce the likelihood of delivery failures, enhance customer satisfaction, and streamline logistics operations. Properly cleansed addresses also contribute to better data analytics and reporting, enabling organizations to make informed decisions based on reliable location data.

Identity resolution

The Redpoint Identity Resolution Template provides configurable match rules:

• Deterministic matching: This method involves the use of exact matches to identify records. Specifically, it looks for matches based on email addresses, CRM IDs, and Loyalty IDs. By relying on precise data points, deterministic matching ensures a high level of accuracy in identifying and merging records, minimizing the risk of errors that can arise from ambiguous data.

• Probabilistic matching: In contrast to deterministic matching, probabilistic matching employs a more flexible approach that combines various data elements, such as name, address, and date of birth (DOB), or name and phone number combinations. This method assesses the likelihood of a match based on the available information, allowing for the identification of records that may not have exact matches but share significant similarities. This approach is particularly useful in scenarios where data may be incomplete or inconsistent.

• Hierarchical resolution: To further refine the matching process, hierarchical resolution introduces source-specific priorities. This strategy helps to prevent false merges by establishing a hierarchy of data sources, ensuring that the most reliable information is given precedence. By applying this method, organizations can enhance the integrity of their data management processes and reduce the likelihood of merging records that should remain distinct.

• Resolution levels: The resolution process can be categorized into different levels: individual, household, and account. Each level represents a different scope of matching, with individual focusing on single entities, household encompassing family units, and account addressing broader organizational or business relationships. By understanding and utilizing these resolution levels, organizations can tailor their data matching efforts to suit their specific needs and objectives, ultimately leading to more effective data management strategies.

The resulting linkages populate the Identity Map table, recording source IDs and confidence scores.

Golden Record construction

After linkage, survivorship logic determines which attributes persist to the Golden Record:

• Rule-based survivorship: Prioritize trusted sources when determining data survivorship. The hierarchy of sources should be established as follows: Customer Relationship Management (CRM) systems take precedence over Point of Sale (POS) systems, which in turn are prioritized over digital sources. This structured approach ensures that the most reliable and validated data is retained, thereby enhancing the integrity of the information.

• Recency-based: To maintain the relevance and accuracy of the data, retain the latest values based on their timestamps. This method ensures that the most current information is always utilized, reflecting the latest interactions and updates. By focusing on recency, organizations can make informed decisions based on the most up-to-date data available.

• Completeness-based: When faced with multiple data entries, it is important to retain the most comprehensive profile. This means selecting data that provides the fullest picture of the individual or entity in question, thereby ensuring that no critical information is overlooked. A complete profile enhances understanding and engagement with customers, leading to better service and relationship management.

• Composite rules: Implementing composite rules allows for the establishment of conditional hierarchies in data management. For instance, it is possible to set source precedence based on specific channels. This flexibility enables organizations to tailor their data retention strategies according to the context and needs of their operations. By allowing for such conditional logic, businesses can optimize their data management processes and ensure that they are always working with the most relevant and reliable information.

Ongoing synchronization

The unified record continuously synchronizes with source systems via incremental loads or event triggers:

• Source deltas automatically refresh the Golden Record, ensuring that the most current and accurate data is always available for decision-making and reporting. This automatic refresh mechanism minimizes manual intervention and reduces the risk of errors, thereby enhancing the overall integrity of the data management process.

• Audit tables retain pre-merge history for traceability. By maintaining a comprehensive log of changes and previous data states, these tables provide transparency and accountability. This historical data is invaluable for audits, allowing stakeholders to track modifications and understand the evolution of the data over time.

• Conflicts are logged and surfaced through data quality dashboards, which serve as critical tools for monitoring data integrity. These dashboards highlight discrepancies and issues in real time, enabling teams to address conflicts swiftly and effectively. By visualizing data quality metrics, organizations can proactively manage data-related challenges, ensuring that the information used for analysis and decision-making is reliable and accurate.

Customization layers

Custom match rules

Organizations extend the base identity rules by:

• Adding custom identifiers such as Loyalty Program ID and Social ID can significantly enhance the tracking and personalization of user experiences. By implementing these unique identifiers, businesses can create tailored interactions that cater to individual preferences and behaviors. This not only fosters a sense of belonging among customers but also facilitates more effective marketing strategies.

• Adjusting fuzzy thresholds for name or address similarity is another crucial step in refining data accuracy. By fine-tuning these thresholds, organizations can improve their ability to match records that may have slight variations, ensuring that data integrity is maintained. This process is essential for minimizing errors in customer databases and enhancing the overall efficiency of data management systems.

• Introducing context-specific match tiers such as Household versus Account can provide a more nuanced approach to data matching. By distinguishing between different levels of relationships, organizations can better understand their customer base and tailor their services accordingly. This strategic differentiation allows for more precise targeting and engagement, ultimately leading to improved customer satisfaction and loyalty.

Attribute extension

The Golden Record template can be expanded with:

• Calculated attributes refer to specific metrics that provide insights into customer behavior and value. Examples include Lifetime Value, which estimates the total revenue a customer will generate over their relationship with a business; Tenure, which measures the length of time a customer has been engaged with the brand; and Channel Engagement, which assesses how actively a customer interacts with various communication channels.

• Aggregated metrics are crucial for understanding overall business performance. Transaction frequency indicates how often customers make purchases, while the Web activity score evaluates the level of engagement a customer has with online platforms. These metrics help businesses identify trends and patterns in customer behavior.

• Derived segments categorize customers based on specific characteristics or behaviors. High-value customers are those who contribute significantly to revenue, while dormant users are individuals who have not engaged with the brand for a certain period. By analyzing these segments, businesses can tailor their marketing strategies and improve customer retention efforts.

Governance and quality monitoring

Customized dashboards track:

• Duplicate rate trends: Analyze the trends in duplicate rates by examining historical data to identify patterns and fluctuations in duplicate occurrences, which can inform strategies to minimize redundancy and enhance data integrity. Monitoring these trends over time allows organizations to implement proactive measures, ensuring that the quality of data remains high and that resources are used efficiently.

• Match rule effectiveness: Evaluate the effectiveness of match rules to optimize data matching processes by assessing how well the established rules identify duplicates and ensure data accuracy. By conducting regular reviews and adjustments of these match rules, organizations can enhance their data quality initiatives. This ongoing evaluation not only improves the precision of data matching but also contributes to better decision-making based on reliable information.

• Data completeness and conformance: Ensure data completeness and conformance to maintain high-quality datasets. Completeness refers to the extent to which all required data is present, while conformance relates to how well the data adheres to predefined standards and formats. Regular audits and validation checks can help identify gaps and inconsistencies in the data, allowing organizations to take corrective actions. By prioritizing data completeness and conformance, businesses can enhance their operational efficiency and improve overall data reliability.

• Source contribution weighting: Understanding source contribution weighting involves analyzing the impact of each source on the overall dataset and determining how much weight should be assigned to the information they provide. By implementing a systematic approach to source contribution weighting, organizations can prioritize high-quality data sources and ensure that their decision-making processes are based on the most reliable information available. This strategic focus on data source evaluation ultimately leads to improved insights and outcomes.

Integration with downstream systems

Golden Records are exposed through:

• Data Services API: This is a powerful tool designed for real-time personalization, enabling businesses to tailor their offerings and communications to individual customer preferences and behaviors. By leveraging this API, organizations can enhance user experiences, increase engagement, and drive conversions through personalized content delivery.

• Data extract templates: These templates serve as essential resources for marketing and analytics tools, including popular platforms such as RPI, Unica, and Snowflake. By utilizing these templates, businesses can streamline their data extraction processes, ensuring that they gather and analyze the most relevant information efficiently. This capability is crucial for making informed decisions based on accurate data insights, ultimately leading to more effective marketing strategies and improved performance metrics.

• Activation frameworks: The activation frameworks are integral for outbound orchestration, exemplified by tools like Redpoint Interaction. These frameworks facilitate the execution of targeted marketing campaigns by coordinating various channels and touchpoints. This orchestration ensures that messages reach the right audience at the right time, maximizing the impact of marketing efforts and fostering stronger customer relationships. By implementing robust activation frameworks, organizations can enhance their marketing agility and responsiveness, adapting quickly to changing market dynamics and customer needs.

Example deployment flow

Best practices

• Leverage templates as accelerators, not replacements. It's essential to customize match and survivorship logic early in the process to ensure that the templates serve as effective tools rather than one-size-fits-all solutions. By doing this, organizations can better align their data management strategies with specific business needs and objectives.

• Implement iterative tuning of match rules through quality assurance (QA) feedback cycles. This approach allows for continuous improvement of the matching process, ensuring that the rules evolve based on real-world performance and feedback. Regularly revisiting and refining these rules can lead to more accurate and reliable data matching.

• Use metadata and audit trails to provide full transparency in merge decisions. By documenting the rationale behind each merge, organizations can maintain accountability and clarity in their data management practices. This transparency is crucial for building trust among stakeholders and ensuring compliance with regulatory requirements.

• Align unification granularity, whether focusing on individual records or households, with marketing and compliance objectives. Understanding the specific goals of marketing campaigns and compliance mandates will guide the decision on how granular the data unification should be. This alignment ensures that the data strategy supports broader organizational aims.

• Automate monitoring through dashboards to maintain Golden Record accuracy over time. Implementing automated systems for tracking data quality and accuracy can help organizations quickly identify discrepancies and take corrective actions. Dashboards provide a visual representation of data health, enabling teams to stay informed and proactive in their data management efforts.