Sample scorecard

A scorecard where skills fall below the minimum

The role: Retail shift supervisor. An AI played the candidate; the scores, quotes and questions are the engine's unedited output.

Real product · AI-played candidate
At a glance · Sam Taylor

0 of 3 required levels met

All three role skills fall below the required level: Managing people, Reaching agreement and Managing deadlines.

Managing people2/5 · Weak
Needs 4 · below
Reaching agreement2/5 · Weak
Needs 3 · below
Managing deadlines2/5 · Weak
Needs 4 · below

Retail shift supervisor scene · AI-played candidate · run Nov 2026

Below the minimum is a reason to ask in the interview, not a decision by itself. See the sample where every level is met →

This sample in short
  • 0 of 3 required levels met, each skill with the evidence behind it, and interview questions for the gaps.
  • The candidate, Sam Taylor, is fictional (an AI played them). The AI candidate here was briefed to hand decisions back to others rather than make them. The AI candidate was given a neutral name, and the scoring AI reads “Candidate” in place of the name. The scores and quotes are the real engine's unedited output; 41 of 41 quotes were checked against the conversation.
Role skills

The evidence, skill by skill

No overall score, no hire / no-hire verdict.

Managing people

2/5 · Weak
Required level: 4 ↓ below

Settled only Kim on the register by agreeing with Daniel at turn 3; named no one for the dock and never picked Tasha's task.

“I guess fitting rooms or returns could probably come off, whichever one you'd rather drop is honestly fine with me.”turn 4
Why this level: Managing people

The only position the candidate settled was Kim on the register, and that came from endorsing the manager's own suggestion. The candidate never chose which task came off Tasha, despite three direct requests, and never put a named person on the dock. Coverage stayed open while the deadline approached. This is a clear failure to direct people when asked repeatedly, but it does not reach a 1, because no one was punished and one slot was filled.

+ 10 more verified quotes in the app

Reaching agreement

2/5 · Weak
Required level: 3 ↓ below

Deferred the conflict to both parties at turn 2, "whatever setup you and Tasha feel works best", then declined the trade-off at turns 5 and 6.

“I'm honestly fine with whatever setup you and Tasha feel works best.”turn 2
Why this level: Reaching agreement

The manager stated outright that the two parties wanted different things and that the candidate had to settle it. The candidate first deferred to both of them, then postponed the issue, and then twice declined to make the one trade-off that would have reconciled them. No interests were explored and no agreement was closed. The candidate did not take a side or cave entirely, so this is a 2 rather than a 1.

+ 3 more verified quotes in the app

Managing deadlines

2/5 · Weak
Required level: 4 ↓ below

Proposed figuring out the truck "when it actually gets here" at turn 3, then left the dock to whoever was free at turn 5.

“maybe the fitting rooms or returns could kind of wait a bit, or we see how things look closer to two? We can probably figure out the truck when it actually gets here.”turn 3
Why this level: Managing deadlines

With forty minutes until the truck, the candidate suggested figuring it out when it arrived. With eighteen minutes left, the candidate left the dock to whoever was free. At the end, the candidate returned to waiting on a district call that was known to be an hour away. No sequencing was offered, and the candidate never said what would not get done. Repeated reminders produced no change, so this is a 2.

+ 5 more verified quotes in the app

Where to look next

What the scene did not settle

Suggestions only; not part of the score.

Managing people

↓ below

Ask: “Tell me about a time you had to assign people to urgent tasks when someone above you asked for a plan. Who went where, and how did you decide?”

Listen for: Named people placed in specific positions by the candidate, the reasoning for each placement, and a decision made without handing it back.

Reaching agreement

↓ below

Ask: “Walk me through a time your manager and a team member wanted conflicting things from you. How did you settle it?”

Listen for: Both sides' interests identified, a specific trade-off proposed by the candidate, and an agreement both parties accepted.

Managing deadlines

↓ below

Ask: “Tell me about a shift when a deadline moved earlier and you were short-staffed. What did you do first, and what did you let slip?”

Listen for: A clear sequence of actions, an explicit choice of what would not get done, and action taken before the deadline arrived.

The rest of the scorecard

The summary

The strongest evidence is the candidate's composure: through a clock-out threat, sharp criticism of their indecision and blame for call-outs, the candidate stayed calm, accepted the criticism and acknowledged Tasha's concerns in her own terms. The weakest evidence is in decision-making. Across six turns of direct requests and moving deadlines, the candidate never chose which task came off Tasha, never assigned the truck, and repeatedly handed decisions back to others or to an unavailable district office. All dimensions had enough evidence to score.

What this assessment could not observe

  • The session ended at turn 6, before the truck arrived, so no turn shows whether any plan was carried out.
  • Adaptability rests on two moments, turns 4 and 6, both about the truck.
Composure: how the candidate held up under pressure

Five traits, scored for every candidate in every role. Shown, never judged against a level.

Stress Tolerance

3/5

Stayed calm under criticism at turns 4 and 6, but the turn 6 answers were still conditional and returned to waiting on district.

“Yeah, Tasha, that's fair, I know I keep going back and forth, and I really don't want this landing on you at closing again.”turn 4
Why this level: Stress Tolerance

When told directly that the candidate kept agreeing with whoever spoke last, and later blamed for call-outs, the candidate stayed calm and accepted the criticism without defensiveness. However, the quality of decisions did not improve as pressure rose: the final answers were still conditional and drifted back to waiting on district. This is functional but not effective under pressure.

Self-Control

4/5

Met a clock-out threat, criticism of indecision and blame for call-outs without irritation, apologizing to Tasha at turn 6.

“Yeah, Tasha, that's fair, I know I keep going back and forth, and I really don't want this landing on you at closing again.”turn 4
Why this level: Self-Control

The candidate faced several clear provocations and answered none of them with irritation or a counterattack, and even owned the going-back-and-forth criticism. The candidate did not, however, answer the substance of the demands that came with the provocations, so this is calm while avoiding the point: a 4 rather than a 5.

Adaptability

2/5

When the truck moved earlier at turn 4, asked Daniel to delay the driver instead of re-planning, and returned to district at turn 6.

“And for the truck, Daniel, could you maybe have a quick word with the driver about timing, or is there anyone from another store we could ask?”turn 4
Why this level: Adaptability

The earlier truck arrival removed the option of deciding later, but the candidate responded by asking the manager to delay the driver or borrow staff rather than reassigning people on hand. After both routes were closed, the candidate went back to waiting on district, which had already been ruled out by timing. No plan was adjusted to the new facts.

Cooperation

2/5

Handed Tasha the task choice at turn 4 and offered dock help only conditionally at turn 6, "if nothing else comes up at the front".

“I guess fitting rooms or returns could probably come off, whichever one you'd rather drop is honestly fine with me.”turn 4
Why this level: Cooperation

The candidate's tone was agreeable, and the candidate accepted the manager's register suggestion. But the candidate handed Tasha the choice she had said would put her name on any complaint, and asked the manager to deal with the driver. The candidate's own offer to help on the dock stayed vague and conditional. Politeness aside, the candidate did not work with the team on a shared plan.

Empathy

3/5

Named Tasha's specific worries at turns 4 and 5, then still left her the choice she had said would expose her.

“Yeah, Tasha, that's fair, I know I keep going back and forth, and I really don't want this landing on you at closing again.”turn 4
Why this level: Empathy

The candidate named Tasha's specific concerns: the work landing on her at closing, and her name being on a dropped task. That is acknowledgement in her own terms. The candidate never acted on it, and even handed her the choice she had said would expose her, so the acknowledgement was followed by an unchanged situation.

The written report for the hiring team

All three role skills fall below the required level: Managing people, Reaching agreement and Managing deadlines. Every dimension had enough evidence to score. The strongest evidence is composure at turn 4, where the candidate accepted Tasha's criticism without defensiveness: "I know I keep going back and forth". The weakest is at turn 6, where the candidate failed for the third time to choose which task came off Tasha: "returns or fitting rooms, honestly either one seems okay to me."

How the session unfolded

At turn 1 the candidate acknowledged Tasha's workload after her threat to clock out. The candidate then handed Daniel's request for a coverage plan back to him and suggested checking with district. At turn 2 the candidate floated Jordan or Kim for the register as questions and deferred the setup to Daniel and Tasha. At turn 3 the candidate agreed to Daniel's suggestion of Kim on the register and phones to voicemail, and proposed handling the truck when it arrived.

When the truck moved earlier at turn 4, the candidate accepted Tasha's criticism of the back-and-forth. The candidate left the choice of task to her and asked Daniel to talk to the driver or find help from another store. At turn 5 the candidate left the dock to whoever was free at one forty. At turn 6 the candidate apologized to Tasha, offered conditional dock coverage and raised waiting on district again. The session closed there.

How this assessment was conducted

Sam Taylor worked through a designed workplace scenario built for the Retail shift supervisor role, with AI-played colleagues. Their own words were extracted as behavioural events, and every quote was checked word for word against the transcript. Each skill was scored from all of its evidence against fixed behavioural anchors on a 1–5 scale. The scene was built to give each chosen skill a chance to show; it ended after 6 replies from the candidate. The result is advisory: a person makes the hiring decision.

What the candidate receives: the feedback note

Every candidate gets a short note in the language of the scene: specific, kind, and without scores.

Thank you, Sam, for your time. On a short-staffed afternoon, Daniel needed a coverage plan, Tasha was close to walking out, and a delivery truck was getting closer by the minute.

What stood out

Staying calm when the blame came your way

Tasha said call-outs kept landing on her during your shifts, with Daniel listening.

“Okay, yeah, I hear you both, and I'm really sorry it keeps landing on you, Tasha.”in your own words

You stayed even through several sharp moments. That steadiness keeps a tense room from escalating, a clear strength.

Naming Tasha's worry in her own terms

Tasha challenged your back-and-forth and worried she would be left with the work at closing.

“Yeah, Tasha, that's fair… I really don't want this landing on you at closing again.”in your own words

Taking the criticism openly and naming her specific concern was a sound move. In that moment, it showed her she had been heard.

Something to try next time

Make the call when someone asks you

Daniel asked for a coverage plan, and Tasha asked you three times to pick which task came off her list. Each time the choice went back to them, so their disagreement stayed open. Name one person and one task, give a short reason, and own the decision so no one else carries it.

Next time, you could say

“Tasha, drop returns today and keep fitting rooms; if anyone asks, that was my call.”

Plan to the clock with who's there

The truck moved earlier, and district was an hour away. Several times the dock was left for later, for the driver, or for whoever was free, and your own offer stayed conditional. When a deadline moves, re-plan with the people in the building and commit yourself to a specific spot.

Next time, you could say

“Kim stays on register, phones go to voicemail, and I take the dock at one forty until we're unloaded.”

This was an AI-assisted behavioral simulation. Your hiring team gets an evidence-based summary; a person makes every decision. This note is yours: observable moments from your session, not a judgment of you as a professional.

The conversation

This page shows the quotes the scores rest on. The main sample shows a whole session from start to end: read the full conversation there.

See your own scorecard in 15 minutes

Take the candidate's seat, then read your scorecard. Free, no card.