We measure behaviour in context, not personality labels.

01 · Source

O*NET workplace skills, not generic "soft skill" buzzwords.

Our catalog draws from the U.S. Department of Labor O*NET database: observable behaviours like Active Listening, Persuasion, and Conflict Resolution. Each skill has a stable ID, a plain-language definition, and a group (Discovery, Influence, Drive, Judgment, Adaptability). We deliberately avoid inventing proprietary trait names that sound scientific but cannot be audited.

When a manager assigns focus skills on a learning path, those IDs flow into simulation config, debrief charts, and dossier exports. A rep sees the same vocabulary in Mirror, Deal IQ, Team IQ, and Transfer. That consistency is what makes cross-sim insights possible.

Situational fit beats self-report: we score what you did in the scene, not what you say you usually do.

02 · Five groups

How we organise the library.

Twenty-four skills sit in five groups. Managers rarely need the full list; they pick the three to five that match the rep's current deal or team craft gap.

All 24 skills, scored in context

Each skill has a method page with wrong-tool moments, scene examples, and how we measure fit in Deal IQ and Team IQ.

03 · Training triangle

Assess, train, measure on both deal craft and team craft.

The same triangle applies to Deal IQ (buyer room) and Team IQ (inside pod). Snapshots diagnose. Full sessions train. Rematch or delta measures uplift.

1. 01

   Assess

   Deal IQ snapshot or Team IQ snapshot on a single episode (often Ep3 retro). ~12-18 minutes. Surfaces gaps before live calls or pod friction.

2. 02

   Train

   Full Deal IQ cycle on a qualified account, or Team IQ arc slice (Season 1 pilot: four episodes). Beats are tagged to skills; rubrics are deterministic.

3. 03

   Measure

   Repeat snapshot on the same scene type, or compare dossier axes week over week. Managers export evidence for QBRs, not vanity completion rates.

04 · Team IQ native axes

Some team signals are composite, not O*NET picks.

Psychological safety contribution, managing up, and peer trust repair are measured by peer and manager reactions in-fiction. They map to library proxies for path routing, but debrief copy names the native axis so coaches know what changed.

• Psychological Safety Contribution

  Peer openness after your beats in pod scenes.

• Managing Up

  Fit signals in manager 1:1 episodes.

• Peer Trust Repair

  Recovery after conflict beats, not avoidance.

• Values-in-Action

  Company values rubric on visible choices.

• Feedback Reception

  Response to coach and manager input in-fiction.

05 · Cross-sim

Customer-strong, team-weak is a real pattern.

Because Deal IQ and Team IQ share focus skill IDs, dossiers can flag mismatches: e.g. strong Active Listening with buyers but mismatch scores when peers vent in a retro forum. That is coaching gold, not a flaw in the rep. It is why we run both simulations on the same enablement spine.

Common questions

Is this a personality assessment?

No. We do not assign types or fixed traits. Focus skills are manager-selected behaviours for the current quarter. Scores reflect situational fit on tagged beats in simulation.

How many focus skills can a manager assign?

Three to five per person. Fewer than three lacks coaching signal; more than five dilutes debrief emphasis.

Who owns the rubrics?

Beat-level rubrics are deterministic (scoreBeat.ts), aligned to softSkillFrameworks.ts. LLMs narrate; they do not own scores.

Where does Team IQ fit vs Deal IQ?

Deal IQ trains qualification and buyer-facing judgement. Team IQ trains inside-pod craft: retro, 1:1, handoff, rumor beats. Same passport, different scenes.