Analytics

Overview

The Analytics section provides deep visibility into your multi-marketplace business performance. It is powered by ClickHouse, a columnar analytics database that delivers sub-second query times even on datasets with millions of orders. All analytics data is isolated per tenant and updated continuously via an ETL pipeline from Supabase.

Key Concepts

ClickHouse — A columnar OLAP database optimized for analytics workloads. NexusCommerce ETLs orders, inventory events, pricing events, and ad spend into ClickHouse where they are aggregated into materialized views for instant dashboard loading.

Materialized Views — Pre-computed aggregations stored in ClickHouse. NexusCommerce maintains materialized views for daily revenue, order counts, return rates, and category performance. These views update incrementally as new data arrives.

ETL Pipeline — Supabase → ClickHouse data pipeline. The pipeline runs continuously, processing new records as they arrive. Typical latency from order placement to analytics visibility: under 5 minutes.

Time Grain — The granularity of time-series charts. Supported: hourly, daily, weekly, monthly. The dashboard selects the appropriate grain automatically based on the selected date range.

Getting Started

Navigate to Analytics in the left sidebar. The dashboard loads immediately using pre-computed materialized views. No configuration is required.

Use the date range picker at the top to select the analysis period. The comparison period (shown as a secondary line on charts) is automatically set to the equivalent prior period.

Features

Revenue Overview

The main revenue chart shows total revenue (sum of order totals) across all marketplaces for the selected period, with:

• Primary line: selected period
• Secondary line (dashed): prior period
• Markers for significant events (new marketplace connections, repricing rule changes)

Under the chart: revenue by marketplace as a stacked bar chart showing relative contribution.

Order Analytics

Order Volume Chart — Total orders per day/week/month with marketplace breakdown.

Average Order Value (AOV) — AOV trend over time. Hover to see per-marketplace AOV.

Order Status Distribution — Donut chart showing completed, in-progress, cancelled, and returned orders.

Fulfillment Method Breakdown — FBM vs. FBA vs. FBS split, with revenue and order count for each.

Category Performance

A breakdown of revenue and order volume by product category. Categories are derived from your product catalog's category assignments.

Table columns:

• Category name
• Revenue (selected period)
• Revenue change vs. prior period
• Order count
• AOV
• Return rate

Click a category to drill down to per-SKU performance.

SKU Performance

The top SKUs table ranks products by revenue for the selected period:

Sortable by any column. Exportable to CSV.

Marketplace Comparison

A side-by-side comparison of all connected marketplaces:

• Revenue
• Order count
• AOV
• Return rate
• Buy Box win rate (where applicable)

Normalized view: see each marketplace's performance as a percentage of total to understand relative contribution.

Revenue Cohort Analysis

Cohort analysis groups customers by their first purchase month and tracks their subsequent purchase behavior. Useful for understanding customer lifetime value (LTV) and retention by marketplace.

Note: cohort analysis is only as rich as the customer identifiers provided by each marketplace. Amazon anonymizes buyer emails, so cohort data for Amazon is approximate.

Custom Reports

Build custom reports by selecting dimensions (marketplace, category, SKU, date) and metrics (revenue, orders, units, returns, margin, ad spend, ROAS).

Reports can be:

• Saved as named reports
• Scheduled for weekly or monthly email delivery
• Exported to CSV or JSON

ClickHouse Query Access

For technical users, direct ClickHouse query access is available via the API:

POST /api/analytics/query
Authorization: Bearer <token>
X-Tenant-ID: <tenant-id>
Content-Type: application/json

{
  "query": "SELECT toStartOfDay(created_at) AS day, sum(total_amount) AS revenue FROM orders WHERE tenant_id = :tenantId AND created_at >= :startDate GROUP BY day ORDER BY day",
  "params": {
    "startDate": "2026-01-01"
  }
}

The tenant_id filter is always enforced server-side regardless of the query contents.

Configuration