YOUR BRIEF
What to work through
Work through an ad serving system as a spoken system design interview. Start by clarifying the scope, sketch a working design and explain the trade-offs behind your choices.
- For each ad slot request, return at most one ad, or none, from the eligible campaigns
- Match targeting: audience segments, keywords and location
- Rank eligible ads by bid and predicted click rate, and run an auction for the price
- Spend each campaign’s budget evenly over its flight, without overspending
- Cap how often one user sees the same ad or campaign
BEFORE YOU START
Constraints to discuss
- An answer within 100 ms end to end, 50 to 70 ms of it on the server
- Up to 1 million ad requests a second at peak in Netflix’s framing
- Worst-case overspend held to a small single-digit percentage
- 99.9% availability per region, serving no ad rather than failing the page
- Spend and impressions reported within 5 minutes
Need a starting point?
Watch for these common gaps as you prepare:
- Querying the campaign database on every ad request instead of an in-memory index
- One central budget counter written on every impression by every server
- Ignoring that spend arrives late, so pacing overshoots at the end of the day
- Running the heavy model on every campaign instead of after retrieval and filters
- Failing the page when the ad path is slow instead of returning no ad
- Putting click counting for reports in the serving path
Why this question is in the collection
These sources connect the problem to company interview reports or preparation guides. They do not guarantee what a company will ask you.
- Netflix: PracHub lists a Netflix onsite from April 2026: an ad pacing system that spreads campaign budgets over their flight while serving live ad requests.
- Netflix: PracHub lists a Netflix onsite from August 2026: a distributed frequency capping service for an ad platform.
- Amazon: PracHub lists an Amazon onsite from April 2026: a low-latency ad server that picks an ad per slot, enforces campaign budgets and tracks impressions and clicks.
- OpenAI: PracHub lists an OpenAI technical screen from September 2025: a multi-region ad serving system that returns targeted ads within 100 ms.
- Reddit: PracHub’s Reddit guide lists an ad-serving API among the onsite design questions candidates report.
- Google: Hello Interview’s community question "Design an Ad Server" lists Google among the companies it was asked at.
- Spotify: Hello Interview’s community question "Design an Ad Server" lists Spotify among the companies it was asked at.