# Structured Questions

> Use structured questions to create binary pass/fail screening criteria for your voice/video interviews. Perfect for verifying specific requirements like work authorization, minimum experience thresholds, or knowledge of required technologies.

## What Are Structured Questions?

Structured questions are a special type of interview question designed for clear, objective evaluation. Unlike traditional interview questions that use 1-5 scoring, structured questions have **binary pass/fail outcomes** based on specific criteria you define.

**Key Differences:**

* ✅ **Binary Evaluation**: Candidates either pass or fail based on specific criteria
* ✅ **No Scoring Range**: No 1-5 scale—just pass/fail
* ✅ **No Score Contribution**: Structured questions do NOT contribute to the candidate's overall interview score
* ✅ **Dealbreaker Capability**: Can automatically disqualify candidates who don't meet requirements
* ✅ **Clear Expectations**: Specific answer formats make evaluation more objective

{% hint style="info" %}
**Important**: Structured questions produce pass/fail results only. They don't affect the candidate's numerical score. Use regular (scored) questions if you want responses to contribute to the overall candidate score.
{% endhint %}

## When to Use Structured Questions

<details>

<summary>✅ <strong>Verification of Basic Requirements</strong></summary>

Perfect for must-have qualifications:

* Work authorization status
* Required certifications or licenses
* Minimum years of experience with specific tools
* Essential skills or knowledge

</details>

<details>

<summary>✅ <strong>Knockout/Dealbreaker Screening</strong></summary>

Use to automatically filter out candidates who don't meet core requirements:

* Must be willing to relocate
* Must have at least 3 years of Python experience
* Must be available for weekend shifts
* Must have specific security clearance

</details>

<details>

<summary>✅ <strong>Technical Knowledge Verification</strong></summary>

Test specific knowledge with defined correct answers:

* Programming language proficiency
* Tool or platform familiarity
* Process or methodology understanding
* Industry-specific knowledge

</details>

{% hint style="info" %}
**When NOT to use**: Structured questions aren't ideal for open-ended discussions, behavioral assessment, or situations requiring nuanced evaluation. For those, use traditional scored questions.
{% endhint %}

## Types of Structured Questions

### Yes/No Questions

Yes/No questions require a simple affirmative or negative answer from the candidate. You define which answer you expect.

**Configuration:**

* **Expected Answer**: Choose "Yes" or "No"
* **Dealbreaker**: Optional—automatically fail candidates who give the wrong answer

**Example Use Cases:**

{% tabs %}
{% tab title="Work Authorization" %}
**Question**: "Do you have valid work authorization in the United States?"

* Expected Answer: Yes
* Dealbreaker: Enabled
* **Why it works**: Clear legal requirement, simple verification
* **Industry**: Applicable to all roles
  {% endtab %}

{% tab title="Physical Requirements" %}
**Question**: "Are you able to lift up to 50 pounds regularly?"

* Expected Answer: Yes
* Dealbreaker: Enabled
* **Why it works**: Essential physical requirement for warehouse/logistics roles
* **Industry**: Light industrial, warehouse, logistics
  {% endtab %}

{% tab title="Availability" %}
**Question**: "Are you available to start within the next two weeks?"

* Expected Answer: Yes
* Dealbreaker: Disabled (nice-to-have, not required)
* **Why it works**: Screens for timeline fit without being eliminatory
* **Industry**: Applicable to all roles
  {% endtab %}

{% tab title="Willingness Requirements" %}
**Question**: "Are you willing to work occasional evenings and weekends?"

* Expected Answer: Yes
* Dealbreaker: Enabled
* **Why it works**: Essential job requirement, clear expectation
* **Industry**: Retail, hospitality, light industrial
  {% endtab %}
  {% endtabs %}

**Best Practices:**

* Keep questions simple and unambiguous
* Only use dealbreaker mode for true must-haves
* Test your phrasing—make sure candidates understand what you're asking
* Consider cultural/language nuances in phrasing

### Minimum Amount Questions

Minimum Amount questions verify that candidates meet numeric thresholds for experience, availability, or other quantifiable criteria.

**Configuration:**

* **Minimum Value**: The threshold number required (e.g., 3, 5, 10)
* **Unit**: Optional descriptor (e.g., "years", "hours per week", "projects")
* **Dealbreaker**: Optional—automatically fail candidates below the threshold

**Example Use Cases:**

{% tabs %}
{% tab title="Warehouse Experience" %}
**Question**: "How many years of experience do you have working in a warehouse?"

* Minimum Value: 3
* Unit: years
* Dealbreaker: Enabled
* **Why it works**: Clear baseline requirement for warehouse roles
* **Industry**: Light industrial, warehouse, logistics
  {% endtab %}

{% tab title="Technical Experience" %}
**Question**: "How many years of experience do you have with Python development?"

* Minimum Value: 3
* Unit: years
* Dealbreaker: Enabled
* **Why it works**: Clear baseline requirement for technical roles
* **Industry**: Software development, engineering
  {% endtab %}

{% tab title="Availability Hours" %}
**Question**: "How many hours per week are you available to work?"

* Minimum Value: 20
* Unit: hours per week
* Dealbreaker: Disabled (preferred but flexible)
* **Why it works**: Checks availability without being eliminatory
* **Industry**: Applicable to all roles, especially part-time positions
  {% endtab %}

{% tab title="Certifications" %}
**Question**: "How many safety certifications do you currently hold?"

* Minimum Value: 2
* Unit: certifications
* Dealbreaker: Enabled
* **Why it works**: Verifies required safety credentials
* **Industry**: Light industrial, manufacturing, construction
  {% endtab %}
  {% endtabs %}

**Best Practices:**

* Be specific about what you're measuring (years, projects, certifications, etc.)
* Consider equivalent experience—your AI interviewer can explore this during follow-ups
* Don't set unrealistic thresholds that might exclude good candidates
* Use units to make the requirement crystal clear

### Multiple Choice Questions

Multiple Choice questions present candidates with a set of options and verify they select the correct answer(s). Each option can include additional context for clarity.

**Configuration:**

* **Options**: Up to 4 answer choices
* **Additional Context**: Optional clarification for each option
* **Correct Options**: Define which option(s) are correct
* **Dealbreaker**: Optional—automatically fail candidates who don't select correct options

**Example Use Cases:**

{% tabs %}
{% tab title="Shift Availability" %}
**Question**: "What shifts are you willing to work?"

* Options:
  * Morning (additional context: "6 AM - 2 PM")
  * Afternoon (additional context: "2 PM - 10 PM")
  * Evening (additional context: "10 PM - 6 AM")
  * Weekends (additional context: "Saturday and Sunday")
* Correct Options: Morning, Evening (or any combination based on needs)
* Dealbreaker: Disabled (allows flexibility in scheduling)
* **Why it works**: Identifies scheduling compatibility, candidates can select multiple shifts
* **Industry**: Light industrial, warehouse, retail, hospitality
  {% endtab %}

{% tab title="Equipment Operation" %}
**Question**: "Which of the following equipment are you certified to operate?"

* Options:
  * Forklift (additional context: "OSHA certified")
  * Pallet Jack (additional context: "Electric or manual")
  * Reach Truck (additional context: "High-reach operations")
  * Order Picker (additional context: "Vertical order picking")
* Correct Options: Forklift, Pallet Jack
* Dealbreaker: Enabled
* **Why it works**: Verifies required equipment certifications for warehouse roles
* **Industry**: Warehouse, logistics, manufacturing
  {% endtab %}

{% tab title="Programming Language Proficiency" %}
**Question**: "Which of the following programming languages have you worked with professionally?"

* Options:
  * Python (additional context: "Used in production environments")
  * JavaScript (additional context: "Frontend or backend development")
  * Java (additional context: "Enterprise applications")
  * Go (additional context: "Backend services or systems programming")
* Correct Options: Python, JavaScript
* Dealbreaker: Enabled
* **Why it works**: Verifies required technical skills clearly
* **Industry**: Software development, engineering
  {% endtab %}

{% tab title="Tools & Platforms" %}
**Question**: "Which database systems are you comfortable working with?"

* Options:
  * PostgreSQL
  * MongoDB
  * MySQL
  * Redis
* Correct Options: PostgreSQL, MongoDB
* Dealbreaker: Disabled
* **Why it works**: Checks for preferred but not required experience
* **Industry**: Software development, data engineering
  {% endtab %}
  {% endtabs %}

**Best Practices:**

* Limit to 4 options maximum to avoid overwhelming candidates
* Use additional context to clarify ambiguous options
* Consider allowing multiple correct answers when appropriate
* Make sure options are mutually exclusive unless intentionally overlapping
* Test the question yourself—ensure options make sense when read aloud

## How to Add Structured Questions

When creating or editing interview questions in your voice/video interview agent:

1. Click **"Add New Question"**
2. Select the **"Structured"** tab (not "Regular")
3. Choose your question type from the dropdown:
   * Yes/No
   * Minimum Amount
   * Multiple Choice
4. Enter your question text
5. Configure the type-specific settings:
   * **Yes/No**: Select expected answer (Yes or No)
   * **Minimum Amount**: Enter minimum value and optional unit
   * **Multiple Choice**: Add options (up to 4) with optional additional context, then mark correct options
6. Toggle **"Mark as dealbreaker"** if this question should automatically disqualify candidates who fail
7. Click **"Save"**

{% hint style="warning" %}
**Important**: Structured questions do NOT have scoring criteria (1-5) or scoring weights. They only have pass/fail evaluation.
{% endhint %}

## Combining Structured and Scored Questions

You can mix structured questions with traditional scored questions in the same interview:

**Example Interview Structure:**

1. **Structured Yes/No** (Dealbreaker): "Are you authorized to work in the US?" → Pass/Fail (no score impact)
2. **Structured Minimum Amount** (Dealbreaker): "How many years of React experience?" → Pass/Fail (no score impact)
3. **Scored Question**: "Tell me about a challenging project you worked on." → 1-5 Score (affects overall score)
4. **Scored Question**: "How do you handle tight deadlines?" → 1-5 Score (affects overall score)
5. **Structured Multiple Choice**: "Which testing frameworks have you used?" → Pass/Fail (no score impact)

**How They Work Together:**

**Structured Questions:**

* Produce only pass/fail results
* Can automatically disqualify candidates (when marked as dealbreakers)
* Do NOT contribute to the candidate's overall numerical score
* Best for binary requirements and knockout criteria

**Scored Questions:**

* Produce 1-5 scores based on your defined criteria
* All scored questions contribute to the candidate's overall score
* Weighted based on score\_weight settings
* Best for evaluating nuanced competencies and skills

This hybrid approach lets you:

* Screen out unqualified candidates early (structured dealbreakers)
* Evaluate nuanced competencies that affect the score (scored questions)
* Verify specific knowledge areas without affecting scores (structured questions)

## Best Practices for Structured Questions

<details>

<summary>✅ <strong>Start with Dealbreakers First</strong></summary>

Place structured dealbreaker questions early in the interview to avoid wasting time if candidates don't meet basic requirements.

</details>

<details>

<summary>✅ <strong>Be Precise with Wording</strong></summary>

Since structured questions drive binary outcomes, clarity is critical:

* Use specific, unambiguous language
* Avoid double negatives
* Test how questions sound when read aloud (they'll be spoken by the AI)
* Consider how non-native speakers might interpret your question

</details>

<details>

<summary>✅ <strong>Don't Overuse Dealbreakers</strong></summary>

Reserve dealbreaker mode for true must-haves:

* Legal requirements (work authorization, licenses)
* Absolute minimums (critical experience thresholds)
* Non-negotiable constraints (schedule, location)

Too many dealbreakers = too few qualified candidates.

</details>

<details>

<summary>✅ <strong>Test Your Questions</strong></summary>

Before going live:

* Complete the interview yourself
* Ask a colleague to test it
* Verify the AI pronounces technical terms correctly
* Ensure your thresholds are realistic

</details>

<details>

<summary>✅ <strong>Use Additional Context Wisely</strong></summary>

For multiple choice questions, use additional context to:

* Clarify ambiguous options
* Provide examples of what you mean
* Help candidates distinguish between similar choices

Don't overdo it—keep context brief.

</details>

## Common Mistakes to Avoid

<details>

<summary>❌ <strong>Setting Unrealistic Thresholds</strong></summary>

**Bad**: "Minimum 10 years of Swift experience" (Swift was released in 2014)

**Good**: "Minimum 3 years of iOS development experience"

Research realistic expectations for your industry and role.

</details>

<details>

<summary>❌ <strong>Ambiguous Yes/No Questions</strong></summary>

**Bad**: "Do you have experience with cloud platforms and also know Docker?"

**Good**: Split into two separate questions:

* "Do you have experience with cloud platforms (AWS, Azure, or GCP)?"
* "Do you have experience with Docker containers?"

</details>

<details>

<summary>❌ <strong>Too Many Correct Options</strong></summary>

**Bad**: "Which 5+ programming languages are you proficient in?" (with 8 marked as correct)

**Good**: "Which of these are your primary programming languages?" (with 2-3 marked as correct)

Be realistic about what "correct" means for your role.

</details>

<details>

<summary>❌ <strong>Using Structured Questions for Nuanced Topics</strong></summary>

**Bad**: "Do you have strong leadership skills?" (Yes/No)

**Good**: Use a scored question: "Tell me about a time when you led a team through a challenge."

Save structured questions for objective, verifiable criteria.

</details>

## Reviewing Structured Question Results

When candidates complete interviews with structured questions, you'll see:

* **Pass/Fail Status**: Clear indication of whether they met the criteria
* **Candidate's Answer**: Their exact response to review
* **Evaluation Details**: AI explanation of why they passed or failed
* **Transcript References**: Links to relevant parts of the conversation

**Important Notes:**

* Structured question results appear separately from the candidate's overall score
* Pass/fail results do NOT affect the numerical score—only scored questions contribute to that
* If a candidate failed a dealbreaker question, they'll be automatically filtered in your dashboard
* You can use pass/fail data for filtering and comparison without it impacting scores

{% hint style="success" %}
**Pro Tip**: Even if you don't use dealbreaker mode, structured questions provide valuable binary data points that make candidate comparison easier. You can quickly filter candidates who met specific requirements without reviewing full transcripts—all without affecting their overall score.
{% endhint %}

## Additional Resources

For additional information not covered here, reach out to <support@heymilo.ai>.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.admin.heymilo.ai/creating-your-interview-agents/how-our-scoring-system-works/structured-questions.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.