- Role
- Product Owner
- Company
- Zalo
- Period
- 02/2023 — 04/2024
- Stack
- A/B testing, Tracking dashboard, Live events, Ad monetization
Context
The job
Zalo Game Center is the games platform inside Zalo, Vietnam's leading messaging app. As Product Owner I owned the roadmap and delivery for one new launch and four live products, working with engineering, design and data teams. My claim here is narrow: I did not build the games or the pipelines, I decided what got measured, tested and run, and in what order.
Insight
The decision: measure first
The tension was simple. Reporting was manual, so every question about the games cost someone a data request, and an ad placement is easy to argue about and hard to settle. I put the tracking dashboard first so that experiments and events would have a scoreboard. Step through the three workstreams and watch the results appear.
The decision
Run experiments first, or build the scoreboard first?
Put the tracking dashboard first, so experiments and events have a scoreboard.
Cost I accepted: Experiments waited until the scoreboard existed.
Source: Thao's CV (Zalo, Feb 2023 to Apr 2024) and her description of the roadmap.
Monetization
Why test at all
Ad placement looks like a design opinion. It is a revenue lever, and at this scale the cheapest way to find out which placement is better is to ask the players. Try the call yourself, then see why a picture is not evidence.
Results
- total ad revenue within six monthsCompany figure
- +30%
- user engagement from monthly eventsCompany figure
- +15%
- paying usersCompany figure
- +3%
What I would do next: write the baseline and the test log down next to every number, so the result can be checked, not just believed.
Source: Thao's CV (Zalo, Feb 2023 to Apr 2024). Baselines are not disclosed, and the figures are shown exactly as written. The −40% manual data extraction from the dashboard is in the walkthrough above.
