Shopify Admin Vip Customer Identifier

Purpose

Ranks customers by lifetime spend and order frequency, identifies the top N% (by value, frequency, or both), and outputs a CSV of VIP candidates. Optionally applies a VIP tag to qualified customers via customerUpdate. Used to build loyalty segments, prioritize white-glove support, or seed exclusive-access campaigns. The lifetime spend and order count are pulled directly from Shopify customer aggregates — no external CRM required.

Prerequisites

• Authenticated Shopify CLI session: shopify store auth --store <domain> --scopes read_customers,read_orders,write_customers
• API scopes: read_customers, read_orders, write_customers (only if tag_customers: true)

Parameters

Safety

⚠️ When tag_customers: true, Step 3 executes customerUpdate mutations that mutate customer tag lists. Tags persist until manually removed. Run with dry_run: true first to confirm the VIP list and qualifying thresholds. The default is dry_run: true — you must explicitly set dry_run: false and tag_customers: true to apply tags.

Workflow Steps

1. OPERATION: customers — query Inputs: first: 250, query: "orders_count:>=<min_orders>", select id, displayName, defaultEmailAddress { emailAddress }, numberOfOrders, amountSpent { amount currencyCode }, tags, pagination cursor Expected output: All customers meeting min_orders threshold; paginate until hasNextPage: false

2. OPERATION: orders — query (only when ranking by frequency, for recency annotation) Inputs: For each top candidate: query: "customer_id:<id>", first: 1, sortKey: CREATED_AT, reverse: true Expected output: Most recent order per candidate to annotate the export

3. Filter to amountSpent.amount >= min_spend. Score each customer: spend → spend; frequency → orders; both → 0.6 × normalized spend + 0.4 × normalized frequency. Take the top top_pct%.

4. OPERATION: customerUpdate — mutation (only if tag_customers: true and dry_run: false) Inputs: input: { id: <customer_id>, tags: [...existing_tags, <tag>] } Expected output: customer.id, customer.tags, userErrors

GraphQL Operations

graphqlCopy code
# customers:query — validated against api_version 2025-01
query VIPCandidateCustomers($first: Int!, $after: String, $query: String) {
  customers(first: $first, after: $after, query: $query) {
    edges {
      node {
        id
        displayName
        firstName
        lastName
        defaultEmailAddress {
          emailAddress
        }
        numberOfOrders
        amountSpent {
          amount
          currencyCode
        }
        tags
        createdAt
      }
    }
    pageInfo {
      hasNextPage
      endCursor
    }
  }
}

graphqlCopy code
# orders:query — validated against api_version 2025-01
query VIPLastOrder($query: String!) {
  orders(first: 1, query: $query, sortKey: CREATED_AT, reverse: true) {
    edges {
      node {
        id
        name
        createdAt
        totalPriceSet {
          shopMoney { amount currencyCode }
        }
      }
    }
  }
}

graphqlCopy code
# customerUpdate:mutation — validated against api_version 2025-01
mutation CustomerUpdateVipTag($input: CustomerInput!) {
  customerUpdate(input: $input) {
    customer {
      id
      displayName
      tags
    }
    userErrors {
      field
      message
    }
  }
}

Session Tracking

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

On start, emit:

Copy code
╔══════════════════════════════════════════════╗
║  SKILL: VIP Customer Identifier              ║
║  Store: <store domain>                       ║
║  Started: <YYYY-MM-DD HH:MM UTC>             ║
╚══════════════════════════════════════════════╝

After each step, emit:

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

If dry_run: true, prefix every mutation step with [DRY RUN] and do not execute it.

On completion, emit:

For format: human (default):

Copy code
══════════════════════════════════════════════
VIP CUSTOMER REPORT
  Customers scanned:    <n>
  Eligible (≥ min):     <n>
  VIPs (top <pct>%):    <n>
  Threshold spend:      $<amount>
  Threshold orders:     <n>
  Customers tagged:     <n>  (or "skipped — dry_run")

  Top 10 VIPs by <rank_by>:
    <name>  Spend: $<amount>  Orders: <n>  Last: <date>
  Output: vip_customers_<date>.csv
══════════════════════════════════════════════

For format: json, emit:

jsonCopy code
{
  "skill": "vip-customer-identifier",
  "store": "<domain>",
  "dry_run": true,
  "rank_by": "spend",
  "outcome": {
    "customers_scanned": 0,
    "eligible": 0,
    "vips_identified": 0,
    "threshold_spend": 0,
    "customers_tagged": 0,
    "errors": 0,
    "output_file": "vip_customers_<date>.csv"
  }
}

Output Format

CSV file vip_customers_<YYYY-MM-DD>.csv with columns: customer_id, name, email, lifetime_spend, currency, orders_count, last_order_date, composite_score, rank, tag_applied

Error Handling

Best Practices

• Use rank_by: both to balance whales with loyalists — pure spend ranking can over-index on one-time large purchases.
• Re-run quarterly with a date-stamped tag (e.g., vip-2026-Q2) so lapsed VIPs roll off rather than accumulating permanently.
• Pair with customer-win-back — VIPs who become inactive should be flagged for high-priority re-engagement.
• Run with dry_run: true first; review the threshold spend value to confirm the cutoff matches your VIP definition.