Model ads demand data for reporting

Q: Model ads demand data for reporting

This is a System Design interview question from Netflix for Software Engineer roles. View the full question and solution on PracHub.

Q: How do I approach System Design interview questions?

System Design questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master system design interviews.

Question

Loading...

You are asked to design the data model (and supporting pipelines at a high level) to represent ads demand for an ads platform.

“Demand” here refers to what advertisers want to buy/spend, and what inventory they are eligible to serve on.

Requirements

Support reporting for:
- Booked budget, remaining budget, and spend to date
- Delivery (impressions, clicks, conversions)
- Pacing vs time (e.g., under/over-delivering)
- Breakdown by advertiser, campaign, ad group/line item, creative, targeting, geo, device, day
Handle campaign lifecycle changes:
- Budget edits, start/end date edits, pause/resume
- Creative swaps and targeting updates
Data sources (assume): impression/click logs (high volume), conversion events, campaign configuration service (slow-changing)
Output: propose a warehouse schema (tables + keys), and outline ETL/ELT steps and key validations.

Deliverables

Core entities and relationships
Fact and dimension tables (grain matters)
How you track slowly changing campaign configuration
Example queries/KPIs the model supports

Model ads demand data for reporting

Requirements

Deliverables

Solution

Comments (0)