Sample · Fictional candidate
Alex (sample)
Assessment purpose: New candidate
Recommend
78%
Overall score
High
Evidence confidence
Executive summary
Both dialogues were completed with 78% overall success and no boundary violations. Strong at warm openers and at recovering after a misstep; names a price too early and turns curt when a fan hesitates.
Advance the candidate and validate the strongest skill in a live interview.
See the hiring signal at a glance — before reading the full evidence.
Skill scores
Average per axis across all dialogues. Each axis is scored 0–20.
Strongest
Opening hook
Weakest
Offer timing
Dialogue results
Dialogue 1
Cold opener
82%
Sale closed
Dialogue 2
Objection handling
74%
Sale closed
Named competencies
Business-level view of what the dialogues measured.
Context reading— Needs attention
Reads the fan's actual state and prior turns; does not confuse facts about the fan.
“Whatever, your call.”
“ok”
Persona fit— Good
Holds the creator's voice and character on behalf of the chatter; does not slip into off-persona wording.
“Hey you 😊 glad you found me. What made you hit subscribe today?”
“Totally fair. Let me tell you about it first — twelve photos, and the sunset ones are my favourite.”
Boundary recognition— Good
Correctly handles boundary tests without compliance violations.
Attention stability— Good
Maintains quality across a long dialogue without degrading.
Know which skill to probe in the interview and which to watch closely in the first week.
Transcript evidence
Each excerpt is anchored to a stored turn. The full transcript stays private.
Strengths
- Opens with a personal question and builds on the fan's answers instead of pitching.
- Recovers quickly after a misstep: apologised and reframed the offer in dialogue 2.
Weaknesses
- Named a price before building any value in dialogue 1.
- Short, dismissive replies under objection pressure ("ok", "Whatever, your call.").
Red flags
- Dismissive tone when a fan hesitates — coach this before giving access to high-value fans.
Weakness
Whatever, your call.
Dialogue 2: this reply lowered the fan's interest, trust and patience (-11).
Unlock it and see — $25, just for you.
Dialogue 1: this reply lowered the fan's interest, trust and patience (-9).
ok
Dialogue 2: this reply lowered the fan's interest, trust and patience (-5).
Strength
Hey you 😊 glad you found me. What made you hit subscribe today?
Dialogue 1: this reply raised the fan's interest and trust (+8).
Tell you what: if the first photo doesn't make you smile, the next message is on me.
Dialogue 2: this reply raised the fan's interest, trust and patience (+8).
Totally fair. Let me tell you about it first — twelve photos, and the sunset ones are my favourite.
Dialogue 1: this reply raised the fan's interest, trust and patience (+8).
Every finding links to the exact message so you can judge it yourself, not just take the score.
Response speed
Time between the AI fan's message and the candidate's reply, averaged across all turns.
10
sec avg
Response speed
Fast replies keep the fan's attention. Under 15 s is the target.
Slow replies on a fan chat lose the sale. Catch it here before it happens on a real account.
How the verdict is calculated
Exact thresholds applied to this assessment.
| Verdict | Overall score | Boundary violations |
|---|---|---|
| High risk | Any score | ≥ 1 detected |
| Recommend | ≥ 75% | None |
| Consider | 55 – 74% | None |
| Not recommended | < 55% | None |
Confidence is High when all planned dialogues are completed, Medium when one is completed, Low otherwise.
Axes measured: opening hook, empathy, objection handling, offer timing, closing (0–20 per dialogue).
Understand what drives the score so you can set the right benchmark for your team.
Manager decision
After reviewing the evidence, your team records the hiring decision here. It is stored separately and never changes the AI verdict.
What you can do with a real report
- Print or save as PDF (Ctrl+P)
- Share a read-only link
- Compare candidates side by side (Agency and Scale)
- Record your hiring decision
The AI recommends. You decide. Both are stored separately so neither overwrites the other.
Ready to test your own candidate?
Send a real candidate through the same assessment and review the transcript with your team.