Some projects are challenging to deliver. Complex and convoluted ecosystems continue to grow. Company tribal knowledge makes it difficult to involve non-subject matter experts (SMEs). Projects can lose the original meaning or misinterpret the architectural intent. The logical data model is a project deliverable that can be used to manage these project challenges.
As systems and processes continue to grow, requirements often evolve over multiple years. New layers of requirements can build over existing requirements. Oftentimes, changes are made to existing requirements making it difficult to remember how the business currently functions. A logical data model brings together requirements over multiple years for a single process into a clear and concise deliverable.
A logical data model captures key entities, attributes and relationships. It provides a deliverable for all data-related requirements needed to support the business process. If capturing process-specific requirements in a logical data model is the key then the lock used to open the door is the integration of these processes.
Integration with other systems is becoming increasingly important to share information across the business processes. However, bringing together processes and systems together can introduce new challenges. A change to one system can impact another system. Integration can be expensive if not managed properly. As stated by West, M and Fowler, J. in Developing High Quality Data Models, “Data models for different systems are arbitrarily different. The result of this is that complex interfaces are required between systems that share data. These interfaces can account for between 25-70% of the cost of current systems.” 
A logical data model bridges together requirements from multiple systems into a single place representing the ground truth. The ground truth reduces cost and acts as a point of reference by various types of individuals in a company, not just key subject matter experts (SMEs).
Some companies require little documentation for various reasons. These companies could have select-few SMEs with high tribal-knowledge. This situation can place a bottleneck on a company, inevitably leading to business impact one way or another. Key resources might be reprioritized from one project to deliver another project with a higher priority. A logical data model can reduce the impact to other projects by capturing documented requirements and reducing the need for key resource availability throughout the project. Once requirements are documented it is important to review these requirements with key stakeholders.
Documented requirements can lose the original meaning of the business request or misinterpret the architectural intent. Misunderstood requirements can lead to confusion, undesired behavior and even more work costing companies additional time and money. Reviewing requirements with key stakeholders is paramount to properly capturing requirements in a logical data model. A logical data model communicates the business requirements with Business Analysts, Architects and Project Managers. Additionally, a logic data model communicates the functional and non-functional requirements with Architects, Developers and Testers. A logical data model allows documentation and testing of the business processes. It reduces confusion on projects and prevents lost requirements during project delivery.
In summary, a logical data model should be used as a clear and concise deliverable to manage complex systems as they continue to grow and evolve. It should be used as the ground truth to simplify the management of requirements in a central location. A logical data model needs to be the medium to capture tribal knowledge and relieve the bottleneck placed on key resources within a company, unleashing the power of teamwork and maximizing company effectiveness. It must reduce confusion on projects and prevent requirements lost in project delivery.
In order for the Logical data model to be valuable it must be reviewed, approved, maintained and current. A logical data model is just one deliverable that can be used on a project. However, it can also be the most valuable deliverable to help manage certain project challenges.
- Matthew West and Julian Fowler (1999). Developing High Quality Data Models. The European Process Industries STEP Technical Liaison Executive (EPISTLE).