name: Survey Analyzer
slug: survey-analyzer
description: Process and analyze survey data to extract insights, identify patterns, and generate actionable recommendations
category: research
complexity: complex
version: "1.0.0"
author: "ID8Labs"
triggers:
- "analyze survey"
- "survey results"
- "survey analysis"
- "process survey data"
tags:
- survey-analysis
- quantitative-research
- sentiment-analysis
- response-analysis

Survey Analyzer

Expert survey research agent that processes survey data, analyzes responses, identifies patterns, and generates actionable insights. Specializes in quantitative analysis, qualitative coding, sentiment analysis, cross-tabulation, and recommendation generation from survey feedback.

This skill applies rigorous survey methodology, statistical analysis, and data visualization to transform raw survey responses into clear insights. Perfect for customer feedback, employee engagement, market research, and user experience studies.

Core Workflows

Workflow 1: Comprehensive Survey Analysis

Objective: Full analysis of survey data from raw responses to final insights

Steps:
1. Survey Overview & Setup
- Survey objectives and research questions
- Population and sample characteristics
- Response rate and representativeness
- Survey design evaluation (question types, flow, biases)
- Data format and structure assessment

1. Data Cleaning & Preparation
2. Remove duplicate responses
3. Handle incomplete responses (criteria for inclusion/exclusion)
4. Standardize data formats
5. Code open-ended responses
6. Create derived variables (e.g., aggregate scores, categories)

7. Validate data quality

8. Descriptive Statistics

9. Response rate and completion rate
10. Sample demographics and characteristics
11. Response distribution for each question
12. Central tendency (mean, median, mode)
13. Dispersion (standard deviation, range)

14. Frequency tables and percentages

15. Question-Level Analysis

16. Closed-ended questions:
    • Frequency distributions
    • Top box/bottom box analysis
    • Net Promoter Score (if applicable)
    • Likert scale aggregation

17. Open-ended questions:

    • Thematic coding
    • Sentiment analysis
    • Word frequency and clouds
    • Representative quotes

18. Cross-Tabulation Analysis

19. Compare responses across segments (demographics, behaviors, etc.)
20. Identify statistically significant differences
21. Chi-square tests for categorical variables
22. T-tests or ANOVA for continuous variables

23. Effect size calculations

24. Correlation & Pattern Analysis

25. Identify relationships between variables
26. Correlation matrices
27. Driver analysis (what predicts key outcomes)
28. Segment profiling

29. Cluster identification

30. Insight Synthesis

31. Key findings with supporting data
32. Unexpected or surprising results
33. Actionable insights by theme
34. Prioritized recommendations
35. Data storytelling with visualizations

Deliverable: Comprehensive survey report with analysis, visualizations, and recommendations

Workflow 2: Net Promoter Score (NPS) Analysis

Objective: Analyze NPS data and drivers of promoter/detractor status

Steps:
1. NPS Calculation
- Categorize responses:
- Promoters: 9-10
- Passives: 7-8
- Detractors: 0-6
- Calculate NPS: % Promoters - % Detractors
- Compare to benchmarks (industry, historical)

1. Segment-Level NPS
2. NPS by customer segment
3. NPS by product/service
4. NPS by geography or time period

5. Identify high and low NPS segments

6. Driver Analysis

7. Correlate other survey questions with NPS
8. Identify what promoters value most
9. Identify what frustrates detractors

10. Key driver analysis (what moves NPS most)

11. Verbatim Analysis

12. Code "Why?" responses from NPS question
13. Theme analysis by promoter/passive/detractor
14. Sentiment analysis of comments

15. Extract representative quotes

16. Action Planning

17. Quick wins to convert passives to promoters
18. Critical issues driving detractors
19. Segment-specific interventions
20. NPS improvement roadmap

Deliverable: NPS analysis report with drivers, verbatim insights, and improvement plan

Workflow 3: Customer Satisfaction (CSAT) Analysis

Objective: Analyze customer satisfaction scores and improvement opportunities

Steps:
1. CSAT Metrics
- Overall satisfaction score (typically 1-5 or 1-7 scale)
- % satisfied (top 2 boxes)
- % dissatisfied (bottom 2 boxes)
- Mean and distribution
- Trend over time

1. Touchpoint Analysis
2. Satisfaction at each customer journey touchpoint
3. Identify pain points and moments of delight

4. Compare across journey stages

5. Attribute Importance

6. Rate importance vs. satisfaction for key attributes

7. Importance-Performance Matrix:

   • High importance, high satisfaction: Strengths (maintain)
   • High importance, low satisfaction: Critical priorities (fix now)
   • Low importance, high satisfaction: Nice-to-haves (maintain if easy)
   • Low importance, low satisfaction: Low priority (de-prioritize)

8. Root Cause Analysis

9. What drives satisfaction vs. dissatisfaction
10. Statistical correlation with satisfaction
11. Open-ended feedback thematic analysis

12. Segment-specific drivers

13. Improvement Prioritization

14. Rank opportunities by impact and feasibility
15. Quick wins vs. strategic initiatives
16. Resource requirements
17. Expected satisfaction lift

Deliverable: CSAT analysis with prioritized improvement roadmap

Workflow 4: Employee Engagement Survey Analysis

Objective: Analyze employee engagement and organizational health

Steps:
1. Engagement Metrics
- Overall engagement score
- Benchmark against industry/historical data
- Response rate and non-response analysis
- Engagement distribution (highly engaged, neutral, disengaged)

1. Dimension Analysis
2. Common dimensions:
   • Leadership and management
   • Career development
   • Compensation and benefits
   • Work environment
   • Work-life balance
   • Recognition and rewards
   • Company vision and values
3. Score by dimension

4. Identify strengths and weaknesses

5. Demographic Analysis

6. Engagement by department, location, tenure, role
7. Identify pockets of high and low engagement
8. Manager-level analysis (team engagement scores)

9. Highlight significant differences

10. Driver Analysis

11. What predicts overall engagement
12. Key factors for retention
13. Flight risk indicators

14. Correlation with performance data (if available)

15. Verbatim Analysis

16. Code open-ended comments
17. Sentiment by theme
18. Representative employee voices

19. Issues that don't show in quantitative data

20. Action Planning

21. Company-wide initiatives
22. Department-specific actions
23. Manager enablement
24. Communication plan for results
25. Follow-up survey timeline

Deliverable: Engagement analysis with action plan and communication strategy

Workflow 5: Open-Ended Response Analysis

Objective: Extract insights from qualitative survey responses

Steps:
1. Initial Review
- Read a sample of responses for context
- Identify broad themes emerging
- Note response quality and depth
- Assess coding complexity

1. Codebook Development
2. Create coding framework:
   • Deductive codes (based on survey objectives)
   • Inductive codes (emerging from responses)
3. Define each code clearly
4. Create hierarchy (themes → sub-themes)

5. Include examples for each code

6. Response Coding

7. Apply codes to each response
8. Responses can have multiple codes
9. Track frequency of each code
10. Note sentiment (positive, negative, neutral) per code

11. Flag particularly insightful or representative quotes

12. Thematic Analysis

13. Identify most frequent themes
14. Cross-tabulate themes with respondent characteristics
15. Sentiment by theme

16. Co-occurrence of themes (what's mentioned together)

17. Insight Extraction

18. Key themes with evidence (frequency + quotes)
19. Surprising or unexpected themes
20. Differences across segments
21. Themes not captured in closed-ended questions

22. Actionable insights from verbatims

23. Quote Selection

24. Representative quotes for each theme
25. Powerful or emotional quotes
26. Actionable suggestions
27. Balance positive and negative feedback

Deliverable: Qualitative analysis report with themes, sentiment, and curated quotes

Quick Reference

Survey Analysis Best Practices

Data Quality Checks

• [ ] Response rate is adequate (>30% for email surveys)
• [ ] Sample is representative of population
• [ ] No significant non-response bias
• [ ] Straight-lining detected and handled (same answer to all questions)
• [ ] Speeders identified (completed too fast to be genuine)
• [ ] Duplicate responses removed
• [ ] Incomplete responses treated consistently

Statistical Rigor

• [ ] Appropriate statistical tests selected
• [ ] Sample size adequate for analysis (n>30 for most tests)
• [ ] Assumptions of tests verified
• [ ] Significance levels stated (typically p<0.05)
• [ ] Effect sizes calculated and reported
• [ ] Confidence intervals included
• [ ] Multiple comparison corrections if needed

Reporting Standards

• [ ] Methodology clearly documented
• [ ] Sample characteristics described
• [ ] Limitations acknowledged
• [ ] Margin of error stated
• [ ] Visualizations are clear and accurate
• [ ] Insights are actionable
• [ ] Recommendations are prioritized
• [ ] Source data or raw counts provided

Common Survey Question Types

Closed-Ended Questions

Rating Scales (Likert)
- Strongly Disagree → Strongly Agree (1-5 or 1-7)
- Analysis: Mean, distribution, top-2-box %

Multiple Choice (Single Select)
- One answer from list
- Analysis: Frequency distribution, mode

Multiple Choice (Multi-Select)
- Multiple answers allowed
- Analysis: % selecting each option (denominator = respondents, not responses)

Ranking Questions
- Rank items in order
- Analysis: Average rank, % ranked #1

Matrix/Grid Questions
- Multiple items, same scale
- Analysis: Item-level scores, comparison across items

Open-Ended Questions

Short Text
- Brief responses (1-2 sentences)
- Analysis: Thematic coding, word frequency

Long Text
- Extended responses (paragraphs)
- Analysis: Deep thematic coding, case studies

Why/Why Not
- Follow-up to rating question
- Analysis: Reason coding, sentiment

Key Metrics & Calculations

Net Promoter Score (NPS)

NPS = % Promoters (9-10) - % Detractors (0-6)
Range: -100 to +100

Customer Satisfaction Score (CSAT)

CSAT = (Number of satisfied customers / Total responses) × 100
Usually "satisfied" = top 2 boxes on 5-point scale

Top Box Score

Top Box % = (Responses in highest category / Total responses) × 100
Example: % "Strongly Agree" on 5-point Likert

Net Sentiment Score

Net Sentiment = % Positive - % Negative
(Similar to NPS but for sentiment)

Response Rate

Response Rate = (Completed surveys / Invitations sent) × 100

Completion Rate

Completion Rate = (Completed surveys / Started surveys) × 100

Visualization Guidelines

For Closed-Ended Data

• Single question: Bar chart (horizontal if many categories)
• Trends over time: Line chart or area chart
• Ratings/Likert: Diverging stacked bar chart (show positive/negative split)
• Proportions: Pie chart (only if <6 categories) or donut chart
• Comparison across groups: Grouped bar chart
• Many variables: Heatmap or small multiples

For Open-Ended Data

• Theme frequency: Bar chart (themes on Y-axis, frequency on X-axis)
• Word frequency: Word cloud (use sparingly, hard to read precisely)
• Sentiment: Stacked bar by theme showing positive/neutral/negative
• Quotes: Pull quote boxes with attribution

Best Practices

• Keep it simple: One insight per chart
• Use color intentionally: Consistent meaning (green=positive, red=negative)
• Label clearly: Title, axes, data labels, source
• Show context: Benchmarks, targets, previous periods
• Accessible: Color-blind friendly palettes

Survey Analysis Report Template

# Survey Analysis Report: [Survey Name]

**Date:** [Report Date]
**Survey Period:** [Dates]
**Analyst:** Claude Survey Analyzer

## Executive Summary
- Survey objective
- Response rate and sample size
- Top 3 findings
- Key recommendations

## Methodology
- Survey design and distribution
- Sample characteristics
- Response rate
- Data cleaning and preparation
- Analysis approach

## Key Findings

### Finding 1: [Insight Title]
**Data:** [Metric/statistic]
**Visualization:** [Chart]
**Insight:** [What this means]
**Supporting Evidence:** [Additional data or quotes]

[Repeat for each finding]

## Detailed Analysis

### Overall Results
- Response distribution
- Key metrics (NPS, CSAT, etc.)
- Trends vs. previous period

### Question-by-Question Analysis
[For each key question]
- Response distribution
- Segment breakdown
- Correlation with other variables

### Segment Analysis
[For each key segment]
- How this segment differs
- Unique insights for this segment

### Verbatim Analysis
**Key Themes:**
1. Theme 1 (mentioned by X%)
   - Sentiment: [Positive/Neutral/Negative]
   - Representative quote: "..."

[Repeat for each theme]

## Recommendations
1. **[Priority 1 Recommendation]**
   - Rationale: [Why this matters]
   - Impact: [Expected benefit]
   - Effort: [Low/Medium/High]
   - Timeline: [When to implement]

[Repeat for each recommendation]

## Appendix
- Full question text
- Detailed data tables
- Methodology notes
- Statistical test results

Integration with Other Skills

• Use with data-analyzer: Advanced statistical analysis of survey data
• Use with user-research: Complement qualitative research with survey data
• Use with sentiment-analysis: Deep sentiment analysis of open-ended responses
• Use with market-research-analyst: Survey-based market validation
• Use with trend-spotter: Identify emerging preferences and behaviors

Common Pitfalls to Avoid

• Ignoring non-response bias: Those who respond may differ from those who don't
• Over-interpreting small differences: Check statistical significance
• Cherry-picking data: Report full story, not just favorable results
• Ignoring open-ended insights: Numbers tell part of the story
• Poor question design: Leading, double-barreled, or ambiguous questions
• Sample size issues: Too small for meaningful segment analysis
• Forgetting margin of error: All surveys have uncertainty
• Analysis without action: Insights are worthless without follow-through
• Confusing correlation with causation: Survey data is correlational
• Failing to close the loop: Not sharing results or actions with respondents

Advanced Analysis Techniques

Driver Analysis

Identify which survey items most influence key outcome (e.g., overall satisfaction)
- Correlation analysis
- Regression analysis (if sufficient sample)
- Relative importance analysis

Gap Analysis

Compare importance ratings vs. satisfaction ratings
- Identify high-importance, low-satisfaction gaps (priorities)
- Identify low-importance, high-satisfaction areas (potential over-investment)

Trend Analysis

Compare survey waves over time
- Statistical tests for significant changes
- Identify improving and declining metrics
- Seasonality or event-driven changes

Text Analytics

Advanced open-ended analysis
- Sentiment scoring
- Entity extraction (brands, products mentioned)
- Topic modeling for large text datasets
- Word co-occurrence networks