Rfm Customer Segmentation

Purpose

Performs full RFM (Recency, Frequency, Monetary) analysis across the entire customer base. Each customer is scored 1-5 on three dimensions — how recently they purchased, how often they purchase, and how much they spend — then classified into actionable segments: Champions, Loyal Customers, Potential Loyalists, At-Risk, Hibernating, and Lost. Read-only — no mutations.

Prerequisites

Parameters

Safety

> ℹ️ Read-only by default. If tag_customers: true, will add tags via customerUpdate mutation — use dry_run: true first.

RFM Segment Definitions

Workflow Steps

Inputs: query: "created_at:>=''", first: 250, select createdAt, totalPriceSet, customer { id, email, firstName, lastName, numberOfOrders }, pagination cursor

Expected output: All orders in window with customer linkage; paginate until complete

Inputs: Customer IDs from each segment for contact details

Expected output: Email, name, tags for top customers in each segment

GraphQL Operations

# orders:query — validated against api_version 2025-01
query OrdersForRFM($query: String!, $after: String) {
  orders(first: 250, after: $after, query: $query) {
    edges {
      node {
        id
        createdAt
        totalPriceSet { shopMoney { amount currencyCode } }
        customer {
          id
          email
          firstName
          lastName
          numberOfOrders
        }
      }
    }
    pageInfo { hasNextPage endCursor }
  }
}

# customers:query — validated against api_version 2025-01
query CustomerDetails($query: String, $after: String) {
  customers(first: 250, after: $after, query: $query) {
    edges {
      node {
        id
        email
        firstName
        lastName
        numberOfOrders
        totalSpentV2 { amount currencyCode }
        tags
        createdAt
      }
    }
    pageInfo { hasNextPage endCursor }
  }
}

Session Tracking

Claude MUST emit the following output at each stage. This is mandatory.

On start, emit:

╔══════════════════════════════════════════════╗
║  SKILL: RFM Customer Segmentation           ║
║  Store: <store domain>                       ║
║  Started: <YYYY-MM-DD HH:MM UTC>             ║
╚══════════════════════════════════════════════╝

After each step, emit:

[N/TOTAL] <QUERY|MUTATION>  <OperationName>
          → Params: <brief summary of key inputs>
          → Result: <count or outcome>

On completion, emit:

For format: human (default):

══════════════════════════════════════════════
RFM SEGMENTATION REPORT  (<days_back> days)
  Customers scored:     <n>
  ─────────────────────────────
  Champions:            <n> (<pct>%)  Avg spend: $<n>
  Loyal Customers:      <n> (<pct>%)  Avg spend: $<n>
  Potential Loyalists:  <n> (<pct>%)  Avg spend: $<n>
  At Risk:              <n> (<pct>%)  Avg spend: $<n>
  Hibernating:          <n> (<pct>%)  Avg spend: $<n>
  Lost:                 <n> (<pct>%)  Avg spend: $<n>

  Top Champions:
    <name> (<email>)  R:<n> F:<n> M:<n>  Spend: $<n>
  Top At-Risk (win-back candidates):
    <name> (<email>)  Last order: <date>  Lifetime: $<n>
  Output: rfm_segments_<date>.csv
══════════════════════════════════════════════

For format: json, emit:

{
  "skill": "rfm-customer-segmentation",
  "store": "<domain>",
  "period_days": 365,
  "customers_scored": 0,
  "segments": {
    "champions": { "count": 0, "pct": 0, "avg_spend": 0 },
    "loyal": { "count": 0, "pct": 0, "avg_spend": 0 },
    "at_risk": { "count": 0, "pct": 0, "avg_spend": 0 },
    "lost": { "count": 0, "pct": 0, "avg_spend": 0 }
  },
  "output_file": "rfm_segments_<date>.csv"
}

Output Format

CSV file rfm_segments_.csv with columns:

customer_id, email, first_name, last_name, recency_days, frequency, monetary, r_score, f_score, m_score, rfm_segment

Error Handling

Best Practices