A simulation is the highest-fidelity assessment in the platform. It puts the assessee in a realistic scenario, presents them with a sequence of choices, and records the path they take. The score is the integral of every decision they made, weighted by what each decision reveals about the competencies the simulation is designed to measure.
What a simulation looks like
A simulation in v0.5 is a branching situational judgement test (SJT). The runner shows the candidate a brief scenario — a few sentences describing a workplace situation — followed by four to six options for what they would do next. Their choice determines the next scene, which presents another set of options, and so on for around five steps.
A typical simulation reads like a short story with reader-chosen turning points. The closer the scenario is to the actual work the role demands, the higher the predictive value of the score.
Ready-made template in v1.0
The platform ships with one ready-to-use simulation template:
- Customer Service Simulation — Frontline agent at the customer line — a 6-node branching scenario placing the assessee in a real situation with an elderly customer upset over a delayed transaction. Measures customer empathy, de-escalation technique, compliance with formal procedure, and timing of escalation to a supervisor.
This simulation is available as a global template in every workspace from the moment it is created, and can be assigned to any employee or candidate from the assessment library.
How a new template gets authored
Simulations in v1.0 are HR-authored with AI assistance. From /{slug}/app/develop/library, click + New simulation. A three-step wizard:
- Pick the role the simulation is targeting. The required competencies of that role become the assessment's target competencies.
- Pick the competencies to focus on — usually 2–4 from the role's required competencies. Trying to assess every competency in one simulation produces shallow scenarios.
- Pick the scenario theme — a difficult stakeholder conversation, a resource-allocation decision, a compliance trade-off, etc.
The model drafts the scenario and the branching tree. You read it, edit any nodes that feel unrealistic, mark the correct and partially correct paths, and publish.
The authoring is non-trivial. A good simulation takes an hour of HR time to refine; a great one takes two. The reward is a re-usable assessment that produces high-quality signal for every assignee.
How the assessee experiences it
Around 15–25 minutes from start to finish. The runner shows one scene at a time; the assessee reads, chooses, and proceeds. There is no back button — the platform is asking what would you do?, and a back button would let the assessee retrace and optimise rather than respond.
Scoring
Each option in each scene has a scoreDelta set during authoring. The total score is the sum of deltas along the path the assessee took. Different paths produce different scores; some paths are explicitly designed as common-but-suboptimal traps, with negative or near-zero deltas.
How the result reaches the competency profile
Simulation results contribute to the simulation's target competencies with fidelity weight 0.9 — the highest in the v0.5 set. The reasoning: the assessee made decisions in a structured scenario, the decisions are observable, and the scoring rubric was authored before the assessee took the test (so it is not post-hoc rationalised).
How to read the report
Three parts:
- The path — the scenes the assessee saw and the option they picked at each, displayed as a tree with their path highlighted.
- The score with per-competency breakdown.
- A narrative that walks through the decisions, calling out where the assessee made strong choices and where they opted for a common trap.
The path view is one of the most useful reports the platform produces, because it shows a manager exactly what their team member would actually do — not what they would say.
When to use it
- Late-stage hiring assessment for senior roles. After the structured interviews, a simulation is the closest the platform gets to try them in the role.
- Promotion decisions, where the question would they make the right call when no one is watching? needs an answer.
- Succession planning, where the bench candidates can be put through the same scenario and ranked by their decisions.
- Leadership development, where a manager can use the simulation report to coach specific decision-making patterns.
When not to use it
- Early in the funnel. Simulations are expensive (in HR's authoring time and the assessee's time). They are not for screening.
- For purely technical competencies. A simulation is a good way to assess judgement and decision-making, not coding ability.
Next
The next pages cover the assessment library — where templates live and how to browse them — and the assignment flow itself.