Academy proposal: observe unaided task completion, retain short evidence-to-decision record, then retest. Separate interface usability from learning gains, market demand and conversion estimates.
Second-edition practitioner guide; contemporary cost and platform assumptions are not preserved as current facts.
Krug recommends early repeated small usability tests centered on meaningful tasks. The annotated Janice demonstration explicitly invents a participant session on a real site; it illustrates facilitation, consent, observation and mental-model mismatch rather than supplying actual study findings.
PDF sheet 195
PDF sheet 157
Usability testing; annotated imaginary session 181–191
Text and rendered pages checked. Verified means faithful paraphrase and locators, not empirical validation of the author's claims. All 252 supplied sheets examined.
Evidence Boundary
The small-round recommendation is formative guidance. The transcript on PDF sheets 181–191 is explicitly imaginary. Recording consent and observer disclosure appear on sheet 183. Do not cite the participant's behavior as real empirical evidence.
Academy Interpretation
Use relevant tasks and participants; record unaided performance and misunderstandings. A successful navigation task does not establish whiskey competence or commercial demand.