Media Planning Migration
Slot-level buying
DSP integration
Cross-media optimization
Overview
From DSP-dependent buying to in-house optimization on Databricks
Out-of-home media buying had been locked into DSP limitations and manual planning for years. We moved the client from fragmented, vendor-dependent workflows to a modern Databricks-powered optimization engine. The new platform handles complex datasets, audience measurements, reach models, quality factors, frame availability, and mobile ad IDs, and transforms them into actionable buying decisions. Now fully integrated into both operational and strategic planning workflows, the transition gave media planners full control over their optimization logic.
Technologies
Challenge
What we faced
- 01Moving complex slot-level buying logic from DSP to in-house systems
- 02Managing intricate selection criteria for including and excluding specific frames
- 03Balancing priorities across POIs, geographic areas, and budget allocations
- 04Bridging the gap between legacy optimization logic and modern platform execution
Approach
How we solved it
- 01Databricks on GCP providing the scalable compute backbone for heavy optimization workloads
- 02GitLab CI/CD ensuring reliable, repeatable deployments across environments
- 03API-first architecture exposing optimization capabilities to planning tools and products
Deliverables
What we built
- 01Transition from legacy DSP workflows to a unified Databricks optimization engine
- 02Seamless integration into operational and strategic planning workflows
- 03Unified data foundation covering audience, reach, quality factors, and availability
- 04API layer enabling product teams and downstream systems to consume optimization results
Ready to leave your legacy behind?
Let's discuss how we can help your team unlock the full potential of AI-powered business intelligence.
Start a conversation