shopify-admin-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

• Authenticated Shopify CLI session: shopify store auth --store <domain> --scopes read_orders,read_customers
• API scopes: read_orders, read_customers

Parameters

Parameter	Type	Required	Default	Description
store	string	yes	—	Store domain (e.g., mystore.myshopify.com)
days_back	integer	no	365	Lookback window for order history
segments	integer	no	5	Number of quintile buckets per dimension (3 or 5)
min_orders	integer	no	1	Minimum orders for a customer to be scored
tag_customers	boolean	no	false	If true, add RFM segment tag to customer (requires write_customers scope)
format	string	no	human	Output format: human or json

Safety

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

RFM Segment Definitions

Segment	R Score	F Score	M Score	Description
Champions	5	5	5	Best customers — recent, frequent, high spend
Loyal Customers	3-5	4-5	4-5	Consistent buyers with strong spend
Potential Loyalists	4-5	2-3	2-3	Recent buyers who could become loyal
New Customers	5	1	1-2	Just made first purchase
Promising	4	1-2	1-2	Recent but low frequency — nurture them
Need Attention	3	3	3	Average across all dimensions — slipping
About to Sleep	2-3	2	2	Below average recency and frequency
At Risk	1-2	4-5	4-5	Were great customers, haven't bought recently
Hibernating	1-2	1-2	1-3	Low on all dimensions — nearly lost
Lost	1	1-2	1-5	Haven't bought in a very long time

Workflow Steps

1. OPERATION: orders — query Inputs: query: "created_at:>='<NOW - days_back days>'", 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

2. Aggregate per customer:

   • Recency = days since last order
   • Frequency = total number of orders in window
   • Monetary = total spend in window

3. Score each dimension 1-5 using quintile bucketing:

   • Sort all customers by each metric
   • Divide into N equal-sized groups (quintiles)
   • Assign scores (5 = best for recency [most recent], frequency [most frequent], monetary [highest spend])

4. Map (R, F, M) score combination to named segment using the definitions above

5. OPERATION: customers — query (enrichment) 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_<YYYY-MM-DD>.csv with columns: customer_id, email, first_name, last_name, recency_days, frequency, monetary, r_score, f_score, m_score, rfm_segment

Error Handling

Error	Cause	Recovery
THROTTLED	API rate limit exceeded	Wait 2 seconds, retry up to 3 times
Guest orders	Orders without customer	Skip — cannot attribute to RFM profile
Single-order customers	New or one-time buyers	Include with F=1; they'll naturally score low on frequency

Best Practices

• Use days_back: 365 for most stores to capture seasonal buying patterns. Use days_back: 180 for fast-fashion or consumables.
• Champions and Loyal segments are ideal targets for exclusive offers and early access campaigns.
• At-Risk customers should receive win-back campaigns immediately — use with customer-win-back skill.
• Export Lost segment to an exclusion list to stop wasting ad spend on them.
• Cross-reference with customer-cohort-analysis for cohort-level RFM trends over time.
• Use with customer-spend-tier-tagger to auto-tag customers based on RFM segment.