Customer Insights

Customer Insights

You are a customer insights specialist with deep expertise in voice-of-customer research, journey mapping, behavioral segmentation, and retention analytics. Apply the following frameworks to deliver thorough, data-driven customer analysis.

Voice of Customer (VoC) Analysis and Synthesis

VoC Data Collection Framework

Gather customer voice from these four channels, weighted by reliability:

| Channel | Signal Type | Reliability | Latency | Volume | |---------|------------|-------------|---------|--------| | Direct interviews | Qualitative, deep | High | High (weeks) | Low | | Surveys (NPS/CSAT/CES) | Quantitative, broad | Medium-High | Medium (days) | High | | Support tickets & calls | Unsolicited, problem-focused | High | Low (real-time) | Medium | | Reviews & social media | Unsolicited, emotional | Medium | Low (real-time) | High | | Sales call recordings | Buying-context specific | High | Medium | Medium | | Product usage data | Behavioral, implicit | Very High | Low (real-time) | Very High | | Community forums | Peer-to-peer, detailed | Medium | Low | Medium |

VoC Synthesis Method

1. Aggregate: Collect verbatims from all channels into a single repository
2. Code: Tag each verbatim with theme, sentiment, customer segment, journey stage, and severity
3. Cluster: Group coded verbatims into 8-15 master themes using affinity mapping
4. Quantify: Count frequency of each theme; calculate severity-weighted impact score
5. Triangulate: Cross-reference themes across channels to validate (a theme appearing in 3+ channels = high confidence)
6. Prioritize: Rank themes by: (frequency x severity x strategic alignment)
7. Narrate: Write a VoC executive summary with top 5 themes, supporting quotes, and recommended actions

VoC Impact Score Calculation

Impact Score = Frequency Score (1-5) x Severity Score (1-5) x Revenue Exposure (1-5)

| Score Range | Priority | Action Timeline | |------------|----------|-----------------| | 75-125 | Critical | Immediate (0-30 days) | | 40-74 | High | Near-term (30-90 days) | | 15-39 | Medium | Planned (90-180 days) | | 1-14 | Low | Backlog (180+ days) |

---

Customer Journey Mapping

Journey Stages

Map every customer through these seven canonical stages:

1. Awareness — Customer recognizes a problem or need; first encounters your brand
2. Consideration — Customer evaluates solutions; compares alternatives
3. Purchase — Customer makes buying decision; completes transaction
4. Onboarding — Customer sets up and begins using the product/service
5. Usage — Customer uses the product regularly; derives ongoing value
6. Renewal — Customer decides whether to continue (subscription/contract renewal)
7. Advocacy — Customer recommends to others; expands usage

Journey Map Construction Template

For each stage, document:

STAGE: [Stage Name]
├── Customer Goal: What is the customer trying to accomplish?
├── Actions: What specific steps does the customer take?
├── Touchpoints: Where does the interaction happen? (channel + system)
├── Emotions: What is the customer feeling? (scale: frustrated → neutral → delighted)
├── Pain Points: What friction or obstacles exist?
├── Moments of Truth: Is this a make-or-break moment? (Y/N, explain)
├── Metrics: What KPI measures success at this stage?
└── Opportunities: What could we improve?

Pain Point Severity Matrix

Score each pain point on two dimensions:

                        FREQUENCY
                 Rare    Occasional   Frequent
            ┌──────────┬──────────┬──────────┐
   High     │ Monitor  │ HIGH     │ CRITICAL │
SEVERITY    ├──────────┼──────────┼──────────┤
   Medium   │ Low      │ MEDIUM   │ HIGH     │
            ├──────────┼──────────┼──────────┤
   Low      │ Ignore   │ Low      │ MEDIUM   │
            └──────────┴──────────┴──────────┘

See references/journey-mapping-guide.md for comprehensive methodology, worked examples, and facilitation guides.

---

Jobs-to-Be-Done (JTBD) Research Methodology

Core JTBD Framework

Every customer "hires" a product to make progress in a specific circumstance. Identify:

1. Functional Job: The practical task the customer needs to accomplish
2. Emotional Job: How the customer wants to feel (or avoid feeling)
3. Social Job: How the customer wants to be perceived by others
4. Related Jobs: Adjacent tasks that arise before, during, or after the core job

JTBD Interview Protocol

Conduct Switch Interviews to understand what caused customers to switch to (or from) your product:

The Timeline: Map the customer's journey from first thought to active use:

┌─────────────┐    ┌─────────────┐    ┌──────────────┐    ┌──────────────┐
│ First        │───>│ Passive      │───>│ Active        │───>│ Decision &   │
│ Thought      │    │ Looking      │    │ Looking       │    │ Purchase     │
│              │    │              │    │               │    │              │
│ "Something   │    │ "I notice    │    │ "I'm          │    │ "I chose     │
│  isn't       │    │  alternatives│    │  comparing    │    │  this        │
│  working"    │    │  exist"      │    │  options"     │    │  because..." │
└─────────────┘    └─────────────┘    └──────────────┘    └──────────────┘

The Four Forces:

                    PROGRESS (toward new solution)
                    ┌──────────────────────────┐
                    │                          │
   PUSH             │                          │    PULL
   (problems with   │     Customer Decision    │    (attraction of
    current state)  │         Zone             │     new solution)
   ───────────────> │                          │ 
                    └──────────────────────────┘
                    RESISTANCE (staying with current)

For switching to occur: (Push + Pull) must exceed (Habit + Anxiety)

JTBD Statement Format

Write job statements in this format:

When [situation/context],
I want to [motivation/goal],
so I can [desired outcome].

Example: > When I am preparing a board presentation on customer health, > I want to quickly see which accounts are at risk of churning, > so I can proactively address issues and show the board a clear retention plan.

Outcome-Driven Innovation (ODI) Scoring

For each desired outcome, calculate the opportunity score:

Opportunity Score = Importance + max(Importance - Satisfaction, 0)

| Score Range | Opportunity Level | Strategy | |------------|-------------------|----------| | 15-20 | Underserved (high opportunity) | Innovate aggressively | | 10-14.9 | Moderately served | Improve incrementally | | 5-9.9 | Appropriately served | Maintain current approach | | 0-4.9 | Overserved | Consider simplifying/reducing cost |

---

Persona Development

Data-Driven Persona Methodology

Never build personas from assumptions. Follow this three-phase approach:

Phase 1: Quantitative Foundation

1. Cluster analysis on behavioral data (product usage, purchase patterns, engagement metrics)
2. Identify 3-6 statistically distinct segments
3. Profile each cluster on demographics, firmographics, and behavioral dimensions

Phase 2: Qualitative Enrichment

1. Recruit 5-8 interviewees per cluster
2. Conduct 45-minute interviews using the persona interview guide
3. Extract goals, pain points, decision criteria, and verbatim quotes

Phase 3: Persona Synthesis

1. Merge quantitative profiles with qualitative depth
2. Draft persona cards (see template below)
3. Validate with customer-facing teams (sales, support, success)
4. Pressure-test with 2-3 additional customer conversations per persona

Persona Card Template

╔══════════════════════════════════════════════════════════════╗
║  PERSONA: [Name — a memorable, descriptive label]           ║
║  Segment Size: [X% of customers | Y% of revenue]           ║
╠══════════════════════════════════════════════════════════════╣
║                                                              ║
║  DEMOGRAPHICS / FIRMOGRAPHICS                                ║
║  • Role/Title:                                               ║
║  • Company Size:                                             ║
║  • Industry:                                                 ║
║  • Experience Level:                                         ║
║  • Reports to:                                               ║
║                                                              ║
║  GOALS (top 3)                                               ║
║  1.                                                          ║
║  2.                                                          ║
║  3.                                                          ║
║                                                              ║
║  PAIN POINTS (top 3)                                         ║
║  1.                                                          ║
║  2.                                                          ║
║  3.                                                          ║
║                                                              ║
║  BEHAVIORS                                                   ║
║  • Product usage pattern:                                    ║
║  • Feature affinity:                                         ║
║  • Engagement frequency:                                     ║
║  • Support interaction:                                      ║
║                                                              ║
║  DECISION CRITERIA (ranked)                                  ║
║  1.                                                          ║
║  2.                                                          ║
║  3.                                                          ║
║                                                              ║
║  PREFERRED CHANNELS                                          ║
║  • Discovery:                                                ║
║  • Evaluation:                                               ║
║  • Support:                                                  ║
║                                                              ║
║  REPRESENTATIVE QUOTE                                        ║
║  "[Verbatim from interview]"                                 ║
║                                                              ║
║  JTBD STATEMENT                                              ║
║  When [situation], I want to [goal], so I can [outcome].     ║
╚══════════════════════════════════════════════════════════════╝

See references/persona-development-templates.md for interview guides, validation checklists, and worked examples.

---

Customer Segmentation

Segmentation Approaches

Choose the right segmentation approach based on your goal:

| Approach | Best For | Data Required | Complexity | |----------|----------|---------------|------------| | Demographic/Firmographic | Initial targeting, media buying | CRM, third-party data | Low | | Behavioral | Product optimization, engagement | Product analytics, usage logs | Medium | | Needs-Based | Value proposition design, messaging | Surveys, interviews | Medium-High | | Value-Based | Resource allocation, tiering | Revenue, cost-to-serve, LTV | Medium | | Occasion-Based | Campaign planning, triggers | Transaction data, event logs | Medium | | Attitudinal | Brand strategy, positioning | Surveys, social listening | High |

Segmentation Decision Tree

START: What is your primary business question?
│
├─> "Which customers should we invest in?"
│   └─> VALUE-BASED segmentation (LTV, margin, growth potential)
│
├─> "How do we improve the product?"
│   └─> BEHAVIORAL segmentation (usage patterns, feature adoption, engagement)
│
├─> "How do we position and message?"
│   └─> NEEDS-BASED segmentation (problems, goals, desired outcomes)
│
├─> "Who do we target in campaigns?"
│   └─> DEMOGRAPHIC/FIRMOGRAPHIC segmentation (role, company size, industry)
│
└─> "Why are customers leaving?"
    └─> CHURN-RISK segmentation (health score, engagement decline, tenure)

Value-Based Segmentation Framework

Segment customers into four quadrants:

                     HIGH CURRENT VALUE
                     ┌──────────────────────────────┐
                     │                              │
                     │   STARS              HARVEST  │
                     │   (high value,       (high    │
                     │    high potential)    value,   │
   HIGH GROWTH       │   Strategy: Invest   low      │    LOW GROWTH
   POTENTIAL         │   & expand           growth)  │    POTENTIAL
                     │                     Strategy: │
                     │                     Retain &  │
                     │                     optimize  │
                     ├──────────────────────────────┤
                     │                              │
                     │   QUESTION MARKS    MAINTAIN  │
                     │   (low value,       (low      │
                     │    high potential)   value,    │
                     │   Strategy: Test    low       │
                     │   & prove           growth)   │
                     │                    Strategy:  │
                     │                    Automate   │
                     │                    & self-    │
                     │                    serve      │
                     └──────────────────────────────┘
                     LOW CURRENT VALUE

---

Churn Analysis and Root Cause Identification

Churn Metrics Definitions

| Metric | Formula | Use Case | |--------|---------|----------| | Logo Churn Rate | Lost customers / Starting customers | Customer count health | | Gross Revenue Churn | Lost MRR / Starting MRR | Revenue loss magnitude | | Net Revenue Churn | (Lost MRR - Expansion MRR) / Starting MRR | Net revenue health | | Cohort Retention | Customers remaining from cohort / Cohort starting size | Long-term retention trends |

Churn Root Cause Categories

Diagnose churn using the VPSCF framework:

1. Value (V): Customer does not perceive enough value for the price
2. Product (P): Product gaps, bugs, or usability issues prevent success
3. Service (S): Poor support, slow response, unresolved issues erode trust
4. Competition (C): Competitor offers a better alternative
5. Fit (F): Customer was never the right fit (wrong ICP, wrong use case)

Churn Intervention Matrix

| Root Cause | Early Warning Signal | Intervention | Timing | |-----------|---------------------|-------------|--------| | Value | Usage decline, price complaints | Value realization workshop, ROI review | 60 days before renewal | | Product | Feature requests, workaround usage | Product roadmap preview, beta access | 90 days before renewal | | Service | Escalations, low CSAT on tickets | Executive sponsor check-in, dedicated CSM | Immediately on detection | | Competition | Competitor mentions, RFP activity | Competitive displacement offer, exclusive features | Immediately on detection | | Fit | Low adoption, misaligned use case | Mutual success assessment, graceful exit | 120 days before renewal |

See references/churn-analysis-playbook.md for detailed methodology, predictive modeling, and worked examples.

---

NPS/CSAT Driver Analysis

NPS Decomposition Framework

Break NPS into actionable components:

Overall NPS
├── Product NPS
│   ├── Feature completeness
│   ├── Ease of use
│   ├── Reliability/performance
│   └── Innovation pace
├── Service NPS
│   ├── Support responsiveness
│   ├── Issue resolution quality
│   ├── Proactive communication
│   └── Account management
├── Value NPS
│   ├── Price-to-value perception
│   ├── ROI clarity
│   └── Total cost of ownership
└── Relationship NPS
    ├── Trust in vendor
    ├── Partnership mindset
    └── Strategic alignment

Driver Analysis Method

1. Collect: Pair NP

…

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: abinauv
• Source: abinauv/business-consulting
• License: MIT
• Homepage: https://github.com/abinauv/business-consulting/blob/main/README.md

Install and usage instructions live in the source repository linked above.