Over-Engineering is Equally as Bad as Under-Engineering
Transformation initiatives are often major shifts in how people work, and the tricky part is finding the right balance when designing a business process or system to avoid over-engineering, as this is equally as bad as under-engineering.
The horrors of under-engineering are often more visible and immediate, as process exceptions occur and teams are frustrated to adopt low-functioning systems; both are bad for morale, especially for those implementing change.
Over-engineering is insidious in both the transformation initiative’s initial delivery timeline and as subsequent changes are made. This is directly due to its delayed onset of effects and often is not immediately recognized as the contributing factor.
Within an initial project, a team often over-indexes on solving more problems or delivering more features than are necessary, slowly increasing complexity along with time spent and budget.
After rolling out changes, the process or system may work seamlessly for main and fringe use cases, so there is no evidence of issues until the organization evolves are additional changes are required.
Then, over-engineering rears its head to make it difficult to unwind specific processes or make underlying technical changes, as complexity has a positively correlated relationship with both time and budget.
Here are a few techniques to prevent over-engineering:
Clearly define success with measurable outcomes and revisit that definition regularly during the initiative to prevent scope creep and scope drift
Set priorities early and aggressively to ensure that stakeholders are aware of what to expect from the initiative and mitigate pushback on deprioritized requirements
Plan to iteratively deliver feedback-driven changes that can emerge as new requirements as the process or system is used or deprioritized requirements that are still necessary and impactful
TLDR: Prevent over-engineering by defining and revisiting the objective, set priorities aggressively, and plan for incremental changes after rollout.