Generated Activity
Synthetic sessions, pageviews, account signups, newsletter signups, and page-level behavior for an imaginary website.
Case Study
Synthetic Web Data Pipeline
A reusable analytics engineering project that generates realistic web analytics data for BI, SQL practice, dbt transformations, dashboarding, and data pipeline demos.
Overview
A realistic analytics sandbox for web data
The project simulates page-level activity for an imaginary website, including sessions, pageviews, signups, newsletter signups, user/device/source behavior, time-based patterns, seasonality, randomness, and configurable behavior through YAML.
Configurable Behavior
Traffic volume, pages, sources, devices, seasonality, growth, randomness, and conversion rules are controlled through project configuration.
Analytics Sandbox
The goal is realistic practice data for dashboards, SQL analysis, dbt transformations, and reporting demos, not fake business claims.
Data generation
How the synthetic data is generated
The generator creates believable analytics records without visiting real websites, scraping, or simulating fake users in a browser.
01
Configuration defines traffic behavior, page structure, event probabilities, seasonality, growth, and randomness.
02
A scheduled Python job generates synthetic daily web analytics records.
03
Generated data is loaded into Postgres at a page-level grain.
04
dbt models transform raw records into reporting-ready analytics tables.
05
The portfolio reads aggregated data server-side and renders a compact live dashboard.
Live dashboard
Generated data snapshot
This dashboard queries the project database server-side and renders only aggregated metrics in the browser.
Web analytics over time
Select a report period and metric to inspect the generated daily data.
Sessions
17,018,002
Pageviews
46,243,601
Signups
585,543
Newsletter signups
415,681
Bounce rate
31.7%
Avg. time
2m 25s
2026-01-01 to 2026-09-12
Total Pageviews: 46,243,601
Daily average Pageviews: 181,347
Last Generated Data
Last generated date: 2026-09-12
Timestamp: Sep 12, 2026, 12:00 AM
Rows generated: 2,250
Status: Success
Test results: Success
Technical stack
Built for analytics engineering practice
The project combines generation, storage, transformation, scheduling, and reporting patterns that mirror real analytics workflows.
PythonPostgresdbt CoreYAML configurationRailway cron/job schedulingSQLAnalytics engineeringDashboarding / BI