Interpreting Results

Interpreting Assessment Results

Learn to extract maximum value from AI agent evaluations and make informed hiring decisions.

Assessment Overview

When you open a completed assessment, you'll see:

• Overall compatibility score - Single number (0-100)
• Agent score breakdown - Individual scores from 6 agents
• Visual score distribution - Chart showing agreement/disagreement
• Summary highlights - Key strengths and concerns
• Detailed agent reports - Full analysis from each agent

Understanding the Overall Score

What It Represents

A weighted average of all six agent scores, showing overall candidate-job fit.

Score Interpretation

• 85-100: Exceptional candidate, strong match
• 70-84: Good candidate, worth interviewing
• 55-69: Moderate fit, proceed with caution
• Below 55: Poor fit, likely not suitable

Important: Don't make decisions on overall score alone—dig into agent details.

When Scores Mislead

High overall, one low agent: Candidate may have a specific weakness that matters for your role.

Moderate overall, high variance: Agents disagree significantly—requires human judgment.

Perfect 100 score: Very rare; scrutinize for potential over-fitting or missing data.

Reading Agent Scores

Agent Score Patterns

Uniform high scores (all 80+):

• Strong candidate across all dimensions
• Low risk hire
• Proceed to interview with confidence

Mixed scores (60-90 range):

• Typical pattern for real candidates
• Focus interview on lower-scoring areas
• Verify strengths highlighted by high-scoring agents

Low Investigator score:

• Resume inconsistencies or red flags
• Verify claims during interview
• May need background check

Low Culture Vulture score:

• Cultural fit concerns
• Team interviews essential
• Consider behavioral questions

Low Devil's Advocate score:

• Significant concerns identified
• Read this agent's report carefully
• Determine if concerns are deal-breakers

Agent Agreement Indicators

Strong consensus (all agents within 10 points):

• High confidence in assessment
• Clear signal (good or bad)
• Trust the evaluation

Moderate variance (15-20 point spread):

• Normal for complex candidates
• Read dissenting opinions
• Use interview to clarify

High variance (30+ point spread):

• Unusual candidate profile
• Agents see very different signals
• Requires deeper human investigation

Detailed Agent Reports

Each agent provides:

Numerical Score

The agent's overall rating (0-100)

Executive Summary

2-3 sentence overview of agent's conclusion

Strengths Identified

Bullet list of top 3-5 positive signals:

• What the agent liked
• Why it matters
• Supporting evidence from resume

Concerns Raised

Bullet list of issues or gaps:

• What's missing or problematic
• Why it's a concern
• Severity level

Supporting Evidence

Specific resume/profile quotes the agent based analysis on

Recommendation

• Strong Hire: Agent strongly recommends
• Hire: Agent recommends, with minor reservations
• Consider: Borderline, interview needed
• Pass: Agent recommends not hiring

Key Insights Section

Auto-generated summary highlighting:

• Top 3 Strengths: Most impressive candidate qualities
• Top 3 Concerns: Most significant issues
• Interview Focus Areas: What to explore in interview
• Consensus Points: What all agents agree on

Comparing to Other Candidates

Assessment page includes:

• Percentile ranking: How this candidate compares to others in campaign
• Similar candidates: Others with comparable scores/profiles
• Score distribution: Where candidate falls in your candidate pool

Use this context to calibrate your interpretation.

Red Flags to Watch For

Investigate further when you see:

• Investigator score below 60
• Devil's Advocate identifies critical issues
• Resume timeline gaps or inconsistencies
• Claims that can't be verified
• Significant culture misalignment

Green Flags to Celebrate

Strong positive signals:

• Pragmatist and Strategist both score 85+
• All agents agree on key strengths
• Culture Vulture identifies strong alignment
• Visionary sees high growth potential

Using Results in Interviews

Structure Interview Based on Assessment

High-scoring areas: Confirm quickly with targeted questions

Low-scoring areas: Deep dive to understand if concerns are valid

Conflicting signals: Ask questions that clarify the disagreement

Sample Interview Planning

If Pragmatist scored low:

• Ask about specific technical projects
• Request detailed explanations of work
• Probe for hands-on experience

If Culture Vulture scored low:

• Ask behavioral questions about work style
• Explore values and preferences
• Discuss team dynamics expectations

Trusting vs. Questioning Results

Trust the AI when:

• Strong consensus among agents
• Assessment aligns with your resume review
• Clear evidence supports conclusions
• Multiple data points confirm signals

Question the AI when:

• Scores seem inconsistent with resume quality
• Very limited information was provided
• Agents show extreme variance
• Gut feeling strongly disagrees

Remember: AI is a tool to inform your decision, not make it for you.

Next Steps

• Explore agent-specific insights for deeper understanding
• Learn assessment patterns across candidates
• Compare candidates side-by-side

Pro Tip

Read Devil's Advocate first: This agent identifies concerns others might miss. If Devil's Advocate finds no major issues, that's a strong signal. If concerns are raised, investigate before proceeding.