Proven Architectures

Deep dives into high-volume, high-complexity measurement systems.

Client

Fever

Role

Head of Data Collection

Designing for Billions: A DWH-First Approach

Challenge

Fever operates in a hyper-growth environment with billions of events. Tool-centric tracking (GA, Mixpanel) created silos and conflicting truths. We needed a single source of truth that was vendor-agnostic.

Solution

I designed a DWH-first architecture where the data warehouse (BigQuery/Snowflake) acts as the primary collector. Events are defined once in a canonical schema and then distributed to downstream tools.

Outcome

A scalable, cost-efficient system where 'Truth' lives in our own infrastructure, not in a third-party tool. Vendor changes became trivial, and data quality became auditable.

Client

Wallapop

Role

Privacy & Governance Architect

Balancing Privacy Compliance with Data Utility

Challenge

Navigating strict GDPR/privacy requirements while maintaining granular product analytics. The friction between legal constraints and data needs was slowing down product insights.

Solution

I architected the integration between the Consent Management Platform (OneTrust) and the Data Customer Data Platform (mParticle). We defined strict governance rules: no data flows unless consent is explicitly verified at the event level.

Outcome

Automated compliance. The system inherently respects user choices without manual intervention, allowing the data team to work fast without risking legal exposure.

Client

Adecco

Role

Measurement Architect

Unified Consent for a Multi-Entity Giant

Challenge

A global group with dozens of brands and websites, each with different consent implementations. Complete lack of visibility and standardization.

Solution

I designed a unified Consent Architecture and implementation guide for all group entities. We standardized the data layer signals for consent across the entire portfolio.

Outcome

Global visibility on compliance. A standardize method that reduced implementation time for new sites from weeks to days.

Client

Astara

Role

Data Quality Architect

Automated Quality Assurance at Scale

Challenge

Managing hundreds of websites where 'tracking breaking' was a daily occurrence. Manual QA was impossible at this scale.

Solution

I built a custom monitoring system that scans data layer pushes and validates them against a central schema. Alerts are triggered instantly when a site deviates from the spec.

Outcome

Proactive quality assurance. We moved from 'finding out tracking is broken 2 weeks later' to 'knowing it broke 5 minutes after deploy'.

08 / Contact

Start a Strategic Conversation

If any of the situations above sound familiar, this is usually the right place to start. Explain your context. I read every message and reply if I can add value.