MDM Selection Criteria

From Glitchdata
Jump to navigation Jump to search

Selection Criteria for Master Data Management solutions can include:

  • User Control - Security
  • User Interface
  • Data Consolidation – connectors, process
  • Data Storage
  • Data Models
  • Data Audit
  • Data Reporting & Analysis
  • Data Distribution – connectors, process
  • Process Management
  • Data Latency
  • Data Archive
  • Multi-Platform
  • Multi-Domain
  • Version Control
  • System Monitoring
  • System Logging
  • System Notification
  • System Performance
  • System Availability
  • Licensing
  • Pricing

User Control

Master data involves access by Data Custodians to maintain and update the “golden” record. Such Data Custodians need to have the authority to update these records and this is set through user control.

User Interface

A User Interface is common in all Master Data systems. It provides a GUI for users to view, update data records, access to additional functionality like data statistics, data quality controls, system logs etc… An intuitive interface provides efficiency to user.

Data Consolidation

Data Consolidation in Master Data Management is frequently an automated process executed by ETLs. Data Consolidation depends on business rules, process and data decision made during Data Arbitrage. The measure of good Data Consolidation capability is by having a broad suite of data connectors, the capability to implement process and data decisions, a dependable ETL development platform.

Data Storage

MDM systems store Master Data records in a form of Data Storage. The Data Storage sub-system can be as simple as a file-base container (eg. H2), to open DB systems (eg. MySQL, PostGres) to enterprise data platforms (eg. Oracle, DB2). The measure of good Data Storage capability is in maintainability, scalability, robustness, performance. A good Master Data Management platform should be flexible enough to be configured for any data storage platform.

Data Audit

Data Audit is the ability to trace the activity that has occurred around a dataset. This is typically focused on actors and actions over a period of time.

Data Reporting & Analysis

Some MDM platforms feature Reporting & Analysis capabilities. This delves into the area of Data Quality, and Intelligence.

Data Distribution

Like Data Consolidation, good Data Distribution capability is the result of having a broad suite of data connectors, the capability to implement process and data decisions, a dependable ETL development platform.

Process Management

Process Management is the ability to fashion workflows and enforce the workflows. (eg. Data approval process). Systems commonly adopt BPMN and BPEL as its underlying platform.

Data Latency

Data Latency measures the speed in which a data change is propagated through the system. Many factors affect data latency ranging from ETL efficiency, infrastructure, scheduling, and process management.

Data Archive

Data Archive represents the ability of a Master Data Management platform to retain historical records of a dataset. The measure of this capability is the inherent capability of the product.

Multi-Platform

This is one of the inclusion criteria for both Forrester and Gartner in selecting MDM contenders. Multi-Platform means that a solution is supported on various platforms like UNIX, Windows. In practice, multi-platform capability is dominated by a preferred platform which is often aligned to the customer base, and the vendor’s suite-of-products.

Multi-Domain

Domain refers to industry domain. (eg. Finance, Phamaceuticals, Oil & Gas, Retail etc…) Many MDM vendors provide off-the-shelf data models for certain domains.

Version Control

Version control is a way to manage the transition of code, data, models across systems. There are 2 levels of version control namely:

  1. Code version management
  2. Data, Models version management

System version management is typically managed by the broader architecture migration planning.

System Monitoring

System monitoring is the ability of a platform to provide information on its state, and to generate alerts if there are issues.

System Logging

System Logging is the capability of a system to log events, errors and other system messages that are important for troubleshooting, auditing, performance measurements etc…

System Notification

System notification is a communications and messaging capability of a platform whereby it sends messages to the administrator, or users to alert them of a situation.

System Performance

System Performance is the ability of a platform to scale, to process more information, or store more data. Metrics can be associated to performance, and its best determined by the actual work done.

System Availability

System availability is the robustness of a platform. This is the ability of a system to remain available and responsive even under load.

Licensing

Most master data management platforms are licensable. They can be licensed by user, by server, by CPU, by core, by data records or combination. Licensing structure can be complex to understand.

Pricing

Like Licensing, pricing follows closely where products can be licensed by user, by server, by CPU, by core, by data records or combination. Pricing may attract volume discounts. Pricing structure can be complex to understand.