Can I vibe code Repeat Customer Insights?

Build a custom Shopify Customer Analytics & Cohort Dashboard using Next.js 14 App Router, TypeScript, Tailwind CSS, Tremor UI, Prisma ORM, and PostgreSQL.

1. DATA MODEL:
- Customer: shopifyCustomerId (unique string), email, firstOrderAt, totalSpent, ordersCount.
- Order: shopifyOrderId (unique string), customerId (relation), totalAmount, netAmount (after discounts/refunds), createdAt, orderNumber.
- RFM score cache: customerId, recencyScore (1-5), frequencyScore (1-5), monetaryScore (1-5), segmentName ('Champion', 'At Risk', 'Loyal', 'Lost').

2. SHOPIFY INTEGRATION:
- Route handler for `/api/webhooks/shopify` supporting HMAC verification for `orders/create`, `orders/updated`, and `orders/cancelled`.
- Worker script to perform initial historical backfill using Shopify Admin GraphQL API Bulk Queries, handling rate limits with exponential backoff.
- Deduplicate events using `shopifyOrderId` and support partial/full refund adjustments on net revenue.

3. ANALYTICAL LOGIC & QUERIES:
- Cohort Retention Matrix: SQL query grouping customers by their initial acquisition month (row) and calculating the percentage who placed a subsequent order in month +N (columns).
- Cumulative LTV Matrix: Average cumulative net spend per customer acquired in Month X at month N.
- RFM Calculation: Classify active customers into quintiles for Recency (days since last order), Frequency (total order count), and Monetary (total net spend), assigning standard segment labels.

4. FRONTEND COMPONENTS:
- Summary KPI cards: Repeat Customer Rate, Average Order Frequency, 30/60/90-Day LTV.
- Retention Heatmap grid rendering retention percentages with background color intensity.
- Interactive RFM Segment breakdown table with filters to export target customer IDs.

5. FAILURE MODES & OUT OF SCOPE:
- Ignore draft orders and unfulfilled pending test orders.
- Handle webhooks arriving out of order by checking order update timestamps.
- Out of scope: Multi-currency conversion (assume store base currency), live ad spend ROI attribution.

$ each button prefixes agent-specific run instructions · build your own product, never copy proprietary code, trademarks or designs