Methodology
Reviewed 2026-09-01
1. Separate exposure from outcome
ILO evidence describes the technical exposure of occupations and tasks. It does not directly measure layoffs, wage compression, or the date a title disappears. Articles preserve that distinction in the opening answer and source notes.
Country, occupation, income-level, and gender differences are used only at the resolution supported by the source. Broad evidence can motivate a question; it cannot prove a local causal story.
2. Audit the real task stack
Role analysis breaks work into inputs, transformations, decisions, outputs, verification, consequences, and accountability. It asks which tasks can be standardized and cheaply checked, and which still depend on context, access, trust, or ownership of a result.
The site favors observable evidence over personality language: workflows improved, exceptions caught, decisions owned, sources reconciled, and recovery handled. Those are portfolio artifacts a reader can build.
3. Mark forecasts as forecasts
Titles such as human escalation designer describe plausible bundles of work, not a verified occupational category or promise of vacancies. Forecast pieces explain the mechanism that could create the role and the existing responsibilities it combines.
Corrections update the page and modified date. Readers can send source or wording challenges to the editorial contact listed on every authority page.
The Task Stack Test
Inputs, each scored from zero to its maximum: How much of your week goes into first drafts, summaries, or standard documents? (0–24); When your work is wrong, who notices first? (0–18); How deep are AI tools already inside your team's core workflow? (0–18); Who has your company hired for your kind of work in the last twelve months? (0–20); If you vanished for a month, what would actually break? (0–20). The five maxima sum to 100.
With a job title selected: final = round(0.55 * quiz_points + 0.45 * role.pressure). Without one: final = quiz_points. The role term is the illustrative pressure value published on that role's page, so the same five answers land differently for different titles. Bands: 0–34 Anchored, 35–54 Drifting, 55–74 Moving, 75–100 Recomposing.
Illustrative score. It reflects task structure, not a prediction about your employer.
