Customer Segmentation Builder

Overview

This skill constructs multi-dimensional customer segments by combining transactional data (RFM), behavioral signals, lifecycle stage, and demographic attributes. It produces named, sized, and actionable segments with tailored marketing strategies, channel preferences, and expected value metrics for CPG and retail e-commerce contexts.

When to Use

• Building or refreshing a customer segmentation model
• Identifying high-value customer cohorts for targeted campaigns
• Developing personalized marketing strategies per segment
• Analyzing customer base composition and health
• Informing product development, assortment planning, or pricing with segment insights
• Creating audience lists for paid media, email, and loyalty programs

Required Inputs

Input	Required	Description
Transaction Data	Yes	Customer-level purchase history: order date, order value, items, frequency
Customer Count	Yes	Total number of unique customers in the dataset
Time Period	Yes	Data range for analysis (minimum 12 months recommended)
Product Categories	Recommended	Category/subcategory taxonomy for purchase behavior analysis
Channel Data	Recommended	Acquisition source, purchase channel (web, app, marketplace, store)
Demographic Data	No	Age, gender, location, household composition
Engagement Data	No	Email opens/clicks, site visits, app sessions, loyalty status
Business Objectives	No	Specific goals (e.g., grow retention, reduce churn, increase AOV)

Methodology

Step 1 — RFM Analysis Foundation

Calculate RFM scores for every customer:

Recency (R): Days since last purchase

R Score 5: Purchased within 0–30 days
R Score 4: Purchased within 31–60 days
R Score 3: Purchased within 61–120 days
R Score 2: Purchased within 121–240 days
R Score 1: Purchased 241+ days ago

Frequency (F): Total number of orders in the analysis period

F Score 5: Top 20% by order count
F Score 4: 60th–80th percentile
F Score 3: 40th–60th percentile
F Score 2: 20th–40th percentile
F Score 1: Bottom 20%

Monetary (M): Total revenue from the customer

M Score 5: Top 20% by total spend
M Score 4: 60th–80th percentile
M Score 3: 40th–60th percentile
M Score 2: 20th–40th percentile
M Score 1: Bottom 20%

Composite RFM Score = concatenation of R, F, M (e.g., "555" = best customers).

Step 2 — Lifecycle Stage Assignment

Map RFM scores to lifecycle stages:

Lifecycle Stage	RFM Pattern	Description
Champions	R:5, F:4–5, M:4–5	Best customers; buy often, spend high, recent
Loyal Customers	R:3–4, F:4–5, M:3–5	Consistent buyers with high frequency
Potential Loyalists	R:4–5, F:2–3, M:2–3	Recent buyers with growth potential
Recent Customers	R:5, F:1, M:1–2	New first-time buyers
Promising	R:3–4, F:1–2, M:1–2	Bought recently, low frequency — need nurturing
Need Attention	R:2–3, F:3–4, M:3–4	Previously loyal, showing decline signals
About to Sleep	R:2, F:1–3, M:1–3	Below average recency, at risk
At Risk	R:1–2, F:4–5, M:4–5	Were high-value buyers, haven't returned
Hibernating	R:1, F:1–2, M:1–2	Low across all dimensions, long inactive
Lost	R:1, F:1, M:1	No activity for extended period

Step 3 — Behavioral Enrichment

Layer behavioral attributes onto lifecycle segments:

Purchase Behavior Patterns:

• Category Loyalists: >70% of purchases in one category
• Cross-Category Shoppers: Purchases across 3+ categories
• Promotion-Sensitive: >50% of orders include a discount code
• Full-Price Buyers: <20% of orders include a discount
• Bulk Buyers: AOV >2× average
• Subscribe & Save: Active subscription orders

Channel Behavior:

• Digital-First: >80% of purchases via web/app
• Omnichannel: Active across digital + physical (if applicable)
• Marketplace-Dependent: Primarily Amazon/Walmart.com purchasers
• Social Commerce: Purchases originating from social platforms

Engagement Behavior:

• Email Engaged: Opens >30%, clicks >5% of emails
• App Active: >4 app sessions/month
• Loyalty Member: Enrolled in loyalty/rewards program
• Review Contributor: Has left 1+ product reviews

Step 4 — Value-Based Tiering

Apply value tiers using the Pareto principle:

Tier	Definition	Typical % of Customers	Typical % of Revenue
Platinum	Top 5% by LTV	5%	25%–35%
Gold	6th–20th percentile	15%	25%–30%
Silver	21st–50th percentile	30%	20%–25%
Bronze	Bottom 50%	50%	10%–20%

Calculate per-tier metrics:

• Average LTV, AOV, purchase frequency, and retention rate
• Contribution margin (revenue minus acquisition and servicing costs)
• Growth trajectory (tier migration rate quarter-over-quarter)

Step 5 — Segment Profile Construction

For each segment, build a comprehensive profile:

Segment: [Name]
├── Size: X,XXX customers (XX% of base)
├── Revenue Share: XX% of total revenue
├── Avg LTV: $XXX | Avg AOV: $XX | Avg Frequency: X.X orders/year
├── Retention Rate: XX%
├── Top Categories: Category A (XX%), Category B (XX%)
├── Channel Mix: Web XX%, App XX%, Marketplace XX%
├── Promotion Sensitivity: Low / Medium / High
├── Preferred Engagement: Email / SMS / Push / Social
├── Lifecycle Stage: [Stage]
├── Value Tier: [Tier]
└── Key Behavioral Tags: [Tag1], [Tag2], [Tag3]

Step 6 — Segment Strategy Mapping

Assign tailored strategies per segment:

Segment	Strategic Objective	Key Tactics	KPI Target
Champions	Maximize value, advocacy	Exclusive access, referral program, VIP perks	Increase AOV 10%, referral rate 15%
Loyal Customers	Deepen relationship	Cross-sell, loyalty tier upgrade, early access	Cross-sell rate 20%, retention 90%+
Potential Loyalists	Accelerate 2nd/3rd purchase	Welcome series, incentivized reorder, bundle offers	2nd purchase rate 40%+ within 60 days
Recent Customers	Convert to repeat	Post-purchase nurture, review request, subscription offer	30-day repeat rate 25%+
Need Attention	Re-engage before churn	Win-back email, special offer, survey	Reactivation rate 15%+
At Risk	Prevent high-value loss	Personal outreach, exclusive discount, feedback request	Save rate 20%+
Hibernating	Selective reactivation	Low-cost reactivation (email only), suppress from paid	Reactivation rate 5%+; suppress if no response
Lost	Suppress or sunset	Remove from active lists; reduce acquisition cost	List hygiene, deliverability improvement

Step 7 — Segment Sizing & Opportunity Analysis

Quantify the business opportunity per segment:

Revenue Uplift Potential = Segment Size × (Target Metric - Current Metric) × AOV

Example:
  Potential Loyalists: 5,000 customers
  Current 2nd purchase rate: 25%
  Target 2nd purchase rate: 40%
  AOV: $45
  Uplift = 5,000 × (0.40 - 0.25) × $45 = $33,750 potential revenue

Rank segments by uplift potential to prioritize investment.

Output Specification

1. Segment Summary Table: All segments with size, revenue share, key metrics, and lifecycle stage
2. Segment Profiles: Detailed profile cards for each segment (per Step 5 format)
3. Strategy Matrix: Per-segment objectives, tactics, channels, and KPI targets
4. Value Distribution Chart: Description of Pareto curve showing customer value concentration
5. Opportunity Sizing: Revenue uplift potential per segment, ranked by priority
6. Implementation Roadmap: Phased plan for activating segments across marketing channels
7. Migration Tracking: Recommended quarterly segment migration analysis framework

Examples

Input: "Build customer segments for our DTC supplements brand. 45,000 customers, 18 months of transaction data, 3 product categories (vitamins, protein, wellness). We have email engagement data."

Output: 8 named segments with full profiles. Champions (4% of base, 28% of revenue) are subscription-heavy, multi-category buyers. Potential Loyalists (18% of base) represent $180K uplift opportunity if 2nd purchase rate increases from 22% to 38%. At Risk segment (6% of base) identified as previously high-value subscription cancellers. Strategy matrix with email sequences, loyalty tier design, and paid media suppression rules.

Input: "Segment our grocery e-commerce customers for a new loyalty program design. 200K customers, want to understand who should be in each loyalty tier."

Output: Four-tier loyalty structure mapped to value tiers with qualifying thresholds. Platinum (top 3%, $2,000+ annual spend) gets free delivery + exclusive products. Gold (next 12%, $800+ annual spend) gets accelerated points. Silver earns standard rewards. Behavioral overlay identifies "deal hunters" within each tier for differentiated offer strategies.

Guidelines

• Use at least 12 months of data to account for seasonal purchase patterns in CPG
• Adjust RFM thresholds to the specific business — a monthly consumable has different frequency benchmarks than a durable good
• Segments must be actionable: if you can't target them in a marketing channel, they're analytical curiosities, not segments
• Aim for 6–10 segments; fewer than 5 lacks granularity, more than 12 is operationally unmanageable
• Always include segment size and revenue share — strategy without sizing is incomplete
• For subscription businesses, factor subscription status as a primary behavioral dimension
• Recommend suppression strategies for unresponsive segments to improve overall marketing efficiency
• Update segments quarterly to capture lifecycle migration and seasonal shifts

Validation Checklist

• RFM scores are calculated with business-appropriate thresholds
• Every customer is assigned to exactly one primary segment
• Segment sizes sum to 100% of the customer base
• Revenue shares are calculated and sum to 100%
• Each segment has a distinct strategic objective and tactic set
• Opportunity sizing is quantified with clear assumptions
• High-value at-risk segments are identified and prioritized
• Segments are platform-activatable (exportable to ESP, ad platforms, CRM)
• Migration tracking framework is defined for ongoing monitoring