Author: Chandresh Lad is a Product Manager at Anaplan.
Building a great model is a significant achievement. Keeping it effective, reliable, and easy to manage as the business evolves is a different challenge.
Many Anaplan implementations begin with a clear objective: deliver a planning solution that meets an immediate business need. The initial project is successful, users adopt the solution, and before long, new requirements and use cases begin to emerge.
From that point, the model continues to evolve. New functionality is introduced, planning processes change, and additional model builders become involved. Without a structured approach to managing that change, even a well-designed model can gradually become more difficult to maintain and more complex to update.
Good model management is a combination of practices that help models remain scalable, reliable, and manageable throughout their lifecycle.
Three areas are particularly important:
- Application Lifecycle Management (ALM)
- Model archiving and deletion
- Consistent operational practices
Together, these provide a strong foundation for maintaining a healthy Anaplan environment.
Application Lifecycle Management: Managing change safely
As models mature, structural changes are inevitable. New modules are added, formulas are updated, and business processes change.
Making those changes directly in a production model can introduce unnecessary risk. Adding a dimension or changing a calculation while users are actively working in the model could impact live data or interrupt a planning cycle.
Application Lifecycle Management (ALM) provides a controlled way to move structural changes from development into production while protecting production data.
A typical model lifecycle might include:
- A Development model, where new functionality and enhancements are built
- A Test or UAT model, where changes can be validated usually at a production model scale
- A Production model, targeted at end users and operating in Deployed Mode
This separation gives teams the opportunity to test changes before they reach end users and reduces the likelihood of unexpected issues in production. It also creates a clearer history of how the model has evolved and where structural changes originated without showing all the changes the model builder made to reach that point in the production model.
ALM can also make collaboration easier. Multiple model builders can work on different enhancements using different development and branching approaches while the production model remains stable and available to users.
ALM is often thought of primarily as a deployment capability, but its value goes further. It provides a framework for managing change in a controlled and repeatable way, allowing models to evolve without disrupting day-to-day operations.
Model archiving and deletion: Keeping workspaces manageable
Knowing when to retire, archive, or delete a model is another important part of model management, but it is often overlooked.
Over time, workspaces can accumulate:
- Legacy project models
- Proof-of-concept models
- Training models
- Historical planning models
- Backup models created during implementation
Even when these models are no longer actively used, they can still consume workspace capacity and make the overall environment more difficult to manage. A regular archiving and deletion process helps keep workspaces organized while ensuring historical information is retained where needed, and permanently removed when it is not.
Before archiving or deleting a model, it is worth considering:
- Is the model still actively used?
- Are users still accessing it?
- Is it needed for audit, regulatory, or compliance purposes?
If the model is no longer required, one option is to move it into a read-only (locked) state before archiving it. This provides a transition period during which users can still access historical information without being able to make changes.
Once a model has been archived and has passed an organization’s retention requirements, it should be permanently deleted. Deleting obsolete models is the final step in workspace hygiene, fully reclaiming capacity and reducing the risk of retaining stale data.
A well-organized workspace is easier to administer and gives teams better visibility of the models needed to support the business.
Practical steps for sustainable model management
Consistent ways of working are what make model management effective over the long term. Shifting focus from how a model is built to how it is administered is critical for long-term success.
Establish clear governance and ownership
Every production model should have clearly defined business and technical owners. Establishing a clear governance process ensures that someone is accountable for reviewing user requests, prioritising updates, and approving changes before they move through the ALM process.
Monitor capacity and performance
Model management requires keeping a proactive eye on the health of your workspace. Administrators should regularly monitor model size, and overall workspace capacity. Tracking these metrics will allow teams to anticipate when archiving, and deletion will be necessary, rather than reacting only when a workspace reaches its limit.
Govern user access proactively
As organizations change, so do user roles. A well-managed model requires regular audits of user access. This includes reviewing assigned roles, verifying Selective Access settings, and identifying inactive users. Routine access reviews not only ensure data security and compliance but also help administrators reclaim unused licenses.
Review models regularly
Regular model reviews can identify opportunities to improve performance and remove functionality that is no longer required. A quarterly hygiene check should look for:
- Obsolete or unused actions
- Historical data that can be cleared
- Unused lists or list properties
- Inefficient import/export processes
Addressing smaller maintenance issues regularly is more manageable than waiting for a point when a major cleanup effort is required.
Treat production as an evolving product
A production model should not be viewed as something that is finished when the implementation project ends. It continues to support the business and needs to evolve as requirements change.
This means considering ongoing planning, change management, performance monitoring, and continuous improvement as part of the model's standard lifecycle.
Takeaway
The success of an Anaplan implementation is not defined only by the quality of the initial solution. It also depends on how effectively the model can adapt as the organization and its requirements change.
Organizations that invest in good model management practices are better placed to introduce new capabilities, reduce risk, and provide a reliable experience for their users.
Building a great model is an important first step. Managing it well is what helps it continue to deliver value over the years ahead.
Questions? Leave a comment!
