 |
| Data Migration Series |
Microsoft made available a module on creating a Data Migration strategy (DMS) for Dynamics 365 solutions [1] and provides a template for the DMS workshop on GitHub. Despite the all-encompassing title, the module focuses mainly on the latter. Here are several points to consider in addition.
The workshop allows to gather most important aspects related to the DM, however it's useful to make the strategy explicit as part of a document and share it with the team. The DMS document should follow the general elaboration of a strategy by providing information about the current and targeted state, goals and objectives, assumptions, dependencies, risks and opportunities, the architecture (including alternatives that need to be explored), direct and interfacing processes, roles involved and their responsibilities, communication paths (including escalation), etc.
Even if the template discusses general aspects, the treatment of best practices is minimal (see some technical and architectural aspects).
The DMS should describe when possible the high-level architecture planned for DM. The architecture section can be updated as the project progresses.
One can plan for 3-4 dry-runs plus the UAT, though further iterations might be needed depending on the data quality, respectively the complexity of the logic needed. Each dry-run will have its own scope (in terms of entities and number of records) and objectives (in terms of passed tests), while the UAT attempts to mimic the same conditions the as DM for Go-Live.
Several iterations are usually needed to identify the entities in scope. Moreover, the scope will suffer several changes as new entities and migration areas are identified. There are high the chances that some data entities will be forgotten or postponed after Go Live. Upon case, the business might approach a phased implementation, each with their own DM,
It's useful to have a list of the entities available in the organization and map them to the entities available in the target system(s). This allows to identify the gaps on the DM side. There will be entities for which data need to be provided, entities for which there's no counterpart in the target system, respectively entities with different levels of detail.
An important step of the process is defining the business rules that apply - what subset of data is migrated (e.g. active referenced master data, open transactions, etc.). The definitions will suffer further changes as business perception about what's needed changes.
As they'll go through multiple changes, it's useful to have such lists in Excel or similar repositories and reference them in the strategy and further documents.
The strategy can also target the post-Go-Live migration activities, though the process and architecture might be slightly different, even if many of the requirements still apply. One can devise the further aspects as part of the DMS or consider them as part of the Data Management strategy and operations.
Despite the considerable overlapping of the DM with Data Management, at least until now the DM is not considered as part of the latter. Big organizations with periodic DMs might need to consider a broader strategy, following to address punctually the deviations from each project.
The DM is usually considered as a subproject of the implementation and should be managed as such. Depending on its scope and characteristics (principles, acceptance criteria) the DM might use a different methodology and timeline than the main project. Defining project's goals/objectives, outcomes, stakeholders, timeline, phases, and milestones, respectively the high-level assumptions, dependencies, risks, opportunities and constraints is Project Management 101.
The DM is critical for an implementation's success. It would be useful to describe the architectural alternatives, as well to provide a series of architectural recommendations.
Previous Post <<||>> Next Post
Resources:
[1] Microsoft learn (2023) Create a data migration strategy for Dynamics 365 solutions (link)
No comments:
Post a Comment