Can I vibe code SegMetrics?

Build a full-funnel marketing attribution backend using Node.js, PostgreSQL, and Express.

1. DATA MODEL:
Create database tables for:
- `visitors` (id, anonymous_id, first_seen_at, initial_utm_source, initial_utm_medium, initial_utm_campaign)
- `identities` (id, visitor_id, email, phone, created_at)
- `events` (id, visitor_id, event_type, url, utm_source, utm_medium, utm_campaign, created_at)
- `orders` (id, visitor_id, email, external_order_id, gross_amount, net_amount, currency, created_at)
- `ad_spend` (id, platform, campaign_id, date, spend, impressions, clicks)

2. TRACKING SCRIPT & INGESTION:
- Create a lightweight JS snippet endpoint `/track.js` that sets a first-party HTTP-only cookie (`_store_vid`) with a 1-year expiration.
- Build an `/ingest` POST endpoint that receives pageview and custom events. Capture current URL, referrer, and all standard UTM parameters (`utm_source`, `utm_medium`, `utm_campaign`, `utm_content`, `utm_term`).
- If a user submits an email (e.g., newsletter signup or checkout start), link that email to their `visitor_id` in the `identities` table.

3. WEBHOOK INGESTION:
- Build a Shopify/Stripe purchase webhook handler. When a purchase occurs, look up the customer by email in `identities`. If matched, link the `orders` record to the existing `visitor_id`.
- If no identity matches, create an unlinked order marked for batch reconciliation.

4. ATTRIBUTION CALCULATIONS:
- Implement three REST API endpoints for reporting:
  a) `/api/reports/first-touch`: Attribute 100% of order value to the first UTM campaign logged for that visitor.
  b) `/api/reports/last-touch`: Attribute 100% of order value to the most recent UTM campaign prior to the order.
  c) `/api/reports/linear`: Distribute order value equally across all distinct UTM campaigns logged for that visitor.
- Implement a cohort LTV API `/api/reports/cohorts` that groups users by acquisition month and calculates 30-day, 60-day, and 90-day cumulative revenue.

5. AD SPEND ROI:
- Build a daily cron function that accepts manual JSON payloads or platform CSVs containing daily spend per campaign.
- Compute ROAS = (Attributed Revenue / Total Spend) grouped by campaign and date range.

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