Assessment scenarios

AI fluency assessments built around realistic workplace decisions

twoprune currently organizes assessment coverage into eight scenario families across analytical, operational and commercial work.

Each scenario is a governed, task-bounded simulation. Scenario availability does not mean twoprune certifies professional competence, validates every task in a role or predicts future job performance.

Current assessment coverage

The current library groups eight workplace scenarios into four coverage areas. Each category page explains what the simulation can make observable and how its results should be interpreted.

Finance and controls

FP&A Analyst and Accounting and Controls scenarios focused on reconciling financial, commercial, operational and policy evidence.

Explore finance and controls

Operations and delivery

Business Operations Analyst and Project Manager scenarios involving priorities, constraints, dependencies and recovery decisions.

Explore operations and delivery

Research and strategy

Market Research and Insights and Strategy Analyst scenarios requiring source evaluation, synthesis and qualified recommendations.

Explore research and strategy

Go-to-market and customer work

Product Marketing / Go-to-Market and Customer Success Manager scenarios involving market, customer and commercial evidence.

Explore go-to-market and customer work

A consistent assessment format across different work contexts

The assignment, evidence pack and expected deliverable vary by scenario, but the core assessment model remains consistent. Participants work inside a structured desktop environment with the tools and information made available for that task.

The six AI fluency pillars remain fixed. Scenario design determines which evidence opportunities are relevant and which interpretations can be supported; it should not silently create a different construct or universal role score.

Key points

  • Synthetic workplace context

    Assignments and materials are designed to resemble business work without exposing real customer or participant information.

  • Task-specific evidence

    The scenario creates relevant opportunities to frame the problem, review sources, use AI, verify claims and deliver a result.

  • Available workflow observations

    Where captured, source review, AI interactions, revisions and submission activity can support interpretation of the final deliverable.

  • Human-readable report

    The report describes observed evidence, interpretation, confidence, risks, limitations and useful follow-up questions.

Choosing a scenario

The closest job title is not always the right assessment. Scenario fit depends on the actual work decision, materials, expected deliverable and evidence opportunities.

A customer should review whether the assignment resembles the work context they want to understand, whether the supplied materials create the intended ambiguity and whether the report will support a legitimate development or evaluation purpose.

If no current scenario is a good fit, twoprune should not imply that an arbitrary custom assessment is already available. A new scenario requires design, specialist review, testing and an explicit interpretation boundary.

Coverage and calibration limitations

Assessment evidence is bounded to the scenario, time limit, tools and information available during the session. Performance may vary with domain familiarity, accessibility needs, tool familiarity and the quality of the supplied materials.

Benchmarks should be used only when the comparison group and assessment family are sufficiently relevant. Early or small cohorts should be described as directional or unavailable.