The purpose of this article to provide a series of best practices that you can use to assess the current state of your own CRM system, research the best practices for further detail and then make improvements to your deployment.

Application Lifecycle Management

• Documented and agreed procedures for promoting releases to production.
• Sufficient pre-production environments for development and testing.
• Automated build and deployment procedures to achieve continuous integration.
• All artefacts managed in a robust source control tool.

Solution Management

• A clear policy for managing publishers and solutions.
• Unmanaged solutions in production, except for managed solutions purchased from a third-party.

Vision & Release Roadmap

• A clear and succinct vision that defines the scope of the CRM system
• An understandable roadmap that communicates a high-level plan for major releases of new functionality.

Technical Documentation

• Customization standards covering: configuration standards for entities, forms, attributes and relationships; coding standards for naming conventions, comments, web resources, plugins and web services; and system integration standards.
• Clear and complete documentation for all the solution components summarising how the component satisfies a requirement in sufficient detail that the component can be understood, modified or extended, upgraded and supported.

Requirements

• A process architecture diagram that provides an overview of the in-scope business processes to provide context to the requirements.
• A robust requirements management system where all requirements are created and updated throughout the application lifecycle with traceability to test cases and releases.
• Each requirement will specify: a title and description, the business value, acceptance criteria, priority to the stakeholder, stakeholder name, date created and current status.

User Interface

• User interface standards for the CRM site map, forms, and views.
• Review of the standards and the implemented features by a user interface designer.

Data Quality

• A set of agreed data quality standards that define completeness, validity, timeliness and accuracy.
• A named person responsible for data quality.
• Custom code to format field values where necessary.
• Procedures to identify and resolve data sources and records that do not meet the data quality standards.
• Integration with trusted third-party data providers rather than relying on user input to create or update critical records.

Roles & Responsibilities

• Project Sponsor, System Owner, or Product Owner: responsible for securing internal resources (people and budget), setting the vision and approving the roadmap.
• Project Manager or ScrumMaster: responsible for coaching the implementation team and communicating the team's progress.
• Solution Architect: responsible for overall architectural design of the solution's component systems.
• Requirements Manager: responsible for managing user requirements.
• Infrastructure Manager: responsible for server and client infrastructure.
• Build Manager: responsible for application lifecycle management.
• Development Manager: responsible for development standards and code quality.
• Test Manager/Quality Assurance Manager: responsible for identifying defects and ensuring solution quality.
• Training Manager: responsible for user training.

User Training & Adoption

• Role-tailored user training is provided to all new users before they are given access to the production CRM organization.
• On-demand user training is available to all existing users.
• An up-to-date user guide covers the system's features in context of the business processes, policies and procedures. A wiki and videos can be more useful than a document.
• Appropriate measures – such as gamification solutions – have been implemented to measure and encourage user adoption.

Application Support

• A first-line helpdesk with established procedures for managing user's questions and issues.
• CRM-trained second-level support analysts to resolve escalated technical issues.
• Specialist third-level support analysts and/or a support agreement with Microsoft and the other software vendors involved in the solution.

Application Performance

• Procedures and systems in place for  monitoring system health and performance (for on-premise deployments).
• A considered approach to installing update rollups.
• Techniques for optimizing the application, data and client tiers are understood, implemented and periodically reviewed.

Data Model & Data Dictionary

• Documented data model highlighting the critical system and custom entities
• Data dictionary describing all the modified system fields and custom fields. 

Logical Security

• Appropriate business unit structure to satisfy the security requirements.
• Appropriate custom security roles and field security profiles to satisfy the security requirements and that can be understood, modified or extended, and supported.
• System security roles have not been modified.

Quality Assurance

• Test cases written for all requirements.
• Test tools in place for recording observations and defects.
• Testing conducted by developers, test professionals and users.
• Automated testing tools used for smoke and full regression testing.