Case Study
Oracclum — From MVP to Scalable Attribution Platform
How I turned an attribution problem from digital marketing teams into a SaaS product with campaign tracking, funnel analytics, integration diagnostics, and a more reliable event ingestion foundation.
Role
Founder & Software Engineer
Scope
Product, frontend, backend, infrastructure
Timeline
Apr 2024 - Jun 2026
Highlights
- Validated the problem through conversations with digital marketers before building the first MVP.
- Built the initial product end to end: tracking script, ingestion API, customer dashboard, and attribution views.
- Evolved the platform from simple campaign attribution into funnel analytics and integration diagnostics.
- Reworked the event ingestion path as traffic grew, improving reliability without exposing customer data.
Context
Oracclum started from a recurring problem I saw among digital marketing professionals: paid campaigns were generating traffic and sales, but it was hard to know which campaign, ad, page, or funnel step was actually responsible for each result.
The first opportunity was not to build the most complex attribution platform possible. It was to make attribution easier to adopt for marketers who needed practical answers, clearer onboarding, and better support than the tools they were already trying to use.
What I Built
I built the first version as a focused MVP for Taboola campaigns. It included a customer tracking script, event ingestion APIs, storage and reporting flows, and a React dashboard where users could inspect campaign, site, ad, and sales attribution data.
As customers started using the product, the most valuable expansion became funnel visibility. I added funnel-step tracking and reporting so users could see where visitors stopped progressing, which turned Oracclum from a simple attribution dashboard into a diagnostic tool for campaign and funnel performance.
Technical Evolution
The early version prioritized speed of learning. Once campaign volume increased, the system needed a more resilient ingestion path. I redesigned the backend around a Go ingestion service, asynchronous processing, and an indexed reporting model so events could be captured reliably and queried predictably.
I also built self-service integration testing. Users could simulate an ad click, move through their funnel, and see whether each expected tracking event arrived. This reduced silent tracking failures and helped customers audit their own funnel changes without depending on founder-led support every time.
Outcome
Oracclum became a real SaaS product with active users, paid usage, onboarding flows, production support, and a public portfolio reconstruction that demonstrates the engineering shape without exposing credentials, customer data, or proprietary operational details.
The most important result for me was the full ownership loop: identifying a real customer pain, validating it with users, shipping an MVP, supporting customers directly, and evolving the architecture only when the product demanded it.