Effectiveness asks whether the task gets done

What it is

Effectiveness answers a binary question plus a quality follow-up: can the target users complete the task in the target context, and is the result correct—completion rate, error counts, error severity, result quality. It does not ask how long it took or how it felt, only whether the thing got done.

Why it happens

Effectiveness is determined by the task's success criteria: the more precisely the criteria are defined, the more stable the measurement. "Completed the task" must be refined into a checkable result state—an order submitted with the right amount, a document saved in the right format—or completion cannot be judged. Error tiers matter because recovery costs differ: an instantly undoable mistake and one that loses data should not share a column.

Studying it

Measure per task: predefine success criteria and an error-coding sheet, observe or log the execution, and report first-attempt completion, final completion, errors per task, and unrecovered errors. Distinguish "eventually completed" from "completed unassisted"; they imply different design work. The coding sheet should separate operational errors, comprehension errors, and system faults—different causes, different fixes.

Where it stops holding

Effectiveness says nothing about cost: 100% completion can come with long detours and heavy undoing. For fault-tolerant tasks like editing, error-rate accounting depends on the task definition; for one-shot tasks like payment, unrecovered errors are what count. The researcher sets the success criteria, and biased criteria manufacture high completion rates.

Applying it

• Write success criteria and an error-coding sheet for every core task before testing begins.
• Set unassisted completion rate as the primary acceptance metric with an explicit threshold.
• Reconcile lab results against production funnel data to find mid-task abandonment invisible in the lab.

Related

• Same group: B5.02.2 Efficiency asks how many resources the effort costs · B5.02.3 Satisfaction concerns the subjective experience · B5.02.4 The three can diverge and must be measured separately
• Nearby: B5.01 Usability Definition · Q2 Usability Testing
• Search terms: effectiveness · task completion · error taxonomy