• (289)-201-3556
  • contact@posidexinsights.com

Core Capabilities

Components of Customer MDM Solutions

Data Loading and Data Integration

  • Seamlessly load data from various sources like Excel, flat files, XML files, relational databases, JSON databases, HDFS, Big Data, and Streaming Data.
  • Automatically discover and acquire metadata from data sources for efficient data management.
  • Intuitive user interface to facilitate easy interaction with metadata for streamlined workflows.
  • Perform custom transformations tailored to specific data processing needs.
  • Split text fields based on delimiters like space and commas for improved data segmentation.
  • Extract, transform, and load data with ease for seamless data integration.
  • Map and rationalize physical data models to logical data models for improved data consistency.
  • Perform basic transformations such as data-type conversions, string manipulations, and simple calculations.
  • Efficiently extract and load large volumes of data to enhance productivity.
  • Create, maintain, and customize data models with configurability, extensibility, and upgradability.
  • Connect and access data stored in various relational DBMS engines like Oracle, IBM DB2, MySQL, and Microsoft SQL Server.
  • Establish connectivity with message queues, including popular middleware products like Oracle AQJMS and Java Messaging Service.
  • Move data in bulk between different data repositories for seamless data management.
  • Acquire data based on time or data-value triggers for real-time insights.
  • Execute data delivery based on event triggers to ensure timely data processing.
  • Schedule data delivery in batch mode or on a predefined schedule for efficient data processing.
  • Capture domain values and create masters for specific attributes to maintain data integrity.
  • Implement predefined and customizable approaches for effective error handling and data quality assurance.
  • Accept data for new insertion, updates, and partial data augmentation to accommodate evolving data requirements.
  • Provide tools and facilities for monitoring and controlling runtime processes for enhanced data management.
 

Data Profiling

  • Perform data profiling, data quality assessment, anomaly detection, and metadata discovery.
  • Utilize prebuilt analyses to examine individual attributes, including minimum, maximum, frequency distributions, and patterns.
  • Identify values that occur frequently and detect outliers or exceptional values.
  • Identify and exclude junk values, generating a cleaning list for data improvement.
  • Access packaged processes for common quality tasks, such as handling incomplete data, resolving conflicts in duplicate records, merging rules, auditing, and more.
  • Present profiling results graphically using various chart formats.
  • Generate textual reports that highlight profiling results for easy understanding.
  • Prebuilt graphical dashboards that display profiling results, including junk values, out-of-format PAN, suspicious DOBs, and more.
  • Schedule the execution of profiling processes using built-in or third-party scheduling functionality.
  • Access standard reports that provide comprehensive visibility into profiling results and data quality metrics.
  • Perform efficient parsing operations to extract and manipulate data elements.
...
...

Data Cleansing and Standardisation

  • Perform basic transformations: convert data types, split strings, concatenate values.
  • Execute advanced transformations: complex parsing tasks.
  • Validate pin codes using Pincode Data.
  • Validate phone numbers/mobile numbers using standard specifications.
  • Customize and extend transformations: develop custom logic and leverage packaged transformations.
  • Merge fields to ensure data completeness.
  • Utilize packaged functionality to address specific data quality issues: standardize names, addresses, phone numbers, and merge duplicate records.
  • Split text fields using packaged knowledge bases: match against terms, names, and more.
  • Customize or expand packaged knowledge bases: add terms or create new ones.
  • Apply prebuilt rules for standardization and cleansing: format addresses, phone numbers, and common identifiers like Tax ID numbers.
  • Regular monitoring and updates for dictionaries within the product.
  • Extract and enrich information such as state, district/city, taluk, village, and pincode.
  • Validate and nullify invalid standard identifiers like PAN numbers.
  • Standardize dates across the dataset.
  • Standardize city/district names for consistency.
  • Expand corporate entity acronyms for clarity.
  • Clean and standardize keywords like "public/private limited."
  • Remove noise-contributing and unwanted special characters.
  • Clean excluded values identified through data profiling.
  • Perform extraction and enrichment in real-time and batch mode.

Matching and Clustering

  • Proprietary algorithms (CLIP for Bulk, Prime 360° for Real-time) convert strings to numbers and determine attribute match extent.
  • Robust facilities in batch and real-time modes for cleansing, matching, identifying, linking, and reconciling customer master data from diverse sources, facilitating the creation and maintenance of a comprehensive customer's golden record.
  • Achieve high precision and recall in data matching.
  • Perform matching on all defined attribute combinations to address data inadequacies and optimize recall.
  • Extend clusters by associating them with user-determined properties.
  • Conduct network analysis for deeper insights and connections.
  • Ensure high-performance operations.
  • Address data inconsistencies and nonuniform attribute availability.
  • Support multi-threading for enhanced efficiency.
  • Run all matching rules simultaneously.
  • Utilize clustering to link records belonging to the same entity.
  • Perform extensive linking to achieve comprehensive results.
  • Employ undirected weighted graphs for advanced analysis.
  • Support dual clustering with clusters based on MPC, but prioritize LPC clusters upon manual verification.
  • Classify and grade matches as perfect, authentic, system, MPC, probable, suggestive, referral, or LPC, thereby emphasizing high precision.
...
...

Data Stewardship and Case Management

  • Enable data stewardship to manage customer data across its life cycle and ensure data governance.
  • User-friendly manual remediation in the UI for linking and delinking customer records with complete auditability and record survivability.
  • Implement a maker-checker facility for enhanced data control and validation.
  • Manage user access and roles effectively through user access management and role creation functionalities.
  • Customize the user interface and workflow of the resolution process to align with specific requirements and preferences.

API and Integration Channels

  • Enable seamless integration across multiple modes
  • Web services interfaces developed in a Service-Oriented Architecture (SOA) environment
  • Support for both SOAP and REST services
  • Secure file exchange through SFTP (SSH File Transfer Protocol)
  • Integration at the table level for enhanced data interoperability

Matching Rule Configuration and Survivorship Rule Building

  • User-friendly interface for creating matching rules
  • Support for multiple Matching Rule Profiles (MRP) with the flexibility to choose one before submitting a request. MRP consists of multiple rules with an 'OR' relation.
  • Matching Rules allow for AND/OR operations between attributes.
  • Option to treat an attribute as optional, matching if available, or considering it as a match even if it is 'NULL'.
  • Flexibility to apply multi-value parameters for cross-referencing matching or matching specific types.
  • Adjustable tolerance for each attribute's matching set, allowing approximate matching for attributes like DOB, Contact No, and Identifiers.
  • Variation in matching tolerance can be set for different rules.
  • Ability to search on complete data or subsets of data (Confinement).
  • Confinement can be applied at the rule or MRP level to enforce all rules.
  • Rules to assign preference to the most reliable sources.
  • Dynamic confinement settings can be defined while building the rule or deferred to apply at runtime when the request is posted.
  • Residual attributes can be designated, contributing to match confidence assessment without participating in the matching process.
  • Assign weightages to attributes to calculate match scores effectively.
  • Results can be classified and labeled into different categories based on business rules.
  • Ability to grade match quality for each category.
  • Rank results to prioritize the best matches at the top, with lower ranks indicating higher match quality.
  • Log creation for rule creation activities.
  • Intuitive interface for defining Survivorship rules.
  • Attribute values can be determined based on Survivorship rules, considering factors such as source, timestamp (aging), latest values prevailing over older ones, longest values, maximum, minimum, average, etc.
...
...

Merging and Customers Golden Record Generation

  • Unified customer record derived from multiple source systems for accurate information
  • Application of survivorship rules to establish the definitive golden record
  • Golden record formation based on MPC clusters (Most Probable Clusters)
  • Periodic recasting of the golden record to incorporate incremental data
  • Generation of handoff files to synchronize the golden record with source systems

Customer Master Data Management Tools

  • Prime360 V2.2: Real-time Search and Matching Engine with Relationship Discovery Module for comprehensive customer 360-degree view and identification of both obvious and non-obvious linkages between records.
  • Clip V2.0: Creation of Golden Records and Unique Customer Identification with RCA (Record Consolidation and Aggregation) capabilities for accurate and reliable data management.

Reports

  • Management Information System (MIS) reports
  • Reports on Data Governance
  • Statistical Reports for Data Matching

Deployment and Infrastructure

  • Cloud-based deployment options including Amazon EC2 and Microsoft Azure
  • Software deployment model with hosted off-premises deployment (SaaS)
  • Deployment support for Linux environments
  • Deployment support for IBM infrastructure
  • Deployment support for Solaris
  • Deployment support for Unix-based environments
  • Deployment support for virtualized server environments
  • Deployment support for Windows environments
  • Deployment support for Wintel environments
  • Support for shared and virtualized implementations
  • Traditional on-premises software installation and deployment
  • High Availability (HA) and High Scalability (HS) support