Compare two UI workflows under the same conditions

Define the task in a sentence, the input object, the starting screen, the required role and the visible completion criterion. Record the interface versions or configurations. If those conditions differ, say why and narrow the comparison accordingly.

Distinguish explanatory comparison from measurement. A side-by-side animation can show where actions differ; it does not establish real elapsed time unless the timing was measured under a stated method. Avoid racing edited clips to manufacture a winner.

A fictional review tool offers a list action and a detail-panel action for assigning a request. The example asks only how the selection path differs, not which method is universally better.

• Shared input: R-045, “Sample artwork check,” currently Unassigned; selected owner Example reviewer.
• Shared start: configured reviewer account, same filtered request list, R-045 visible.
• Workflow A: open R-045, open Owner, select Example reviewer, save using the verified detail workflow.
• Workflow B: select R-045 in the list, use the verified assignment control, choose Example reviewer and confirm.
• Shared finish: reopen R-045 and inspect the saved owner. Completion is not merely closing a menu.
• Boundary: the example explains two paths for one assignment; it does not compare setup effort, reliability or all request types.

Align both versions at start, owner selection and confirmed result. Use labels to explain when one path opens a detail panel while the other stays in the list. Let the viewer inspect the difference without forcing equal animation speed.

If one route requires a permission the other does not, that is part of the comparison, not a detail to hide. If both require setup, establish it equally. Never remove an inconvenient confirmation from only one side.

A defensible closing line is “The list route keeps this assignment in the queue.” “Twice as productive” is not supported by the scene count. Even a measured click count would describe this defined task rather than all work.

For an old-versus-new interface comparison, identify the versions and preserve equivalent sample data. For two user preferences, explain when inspecting details is useful rather than assuming fewer visible steps always means a better decision.