WhatsApp call-quality dashboard: who it serves, what it measures, how it is laid out

Quick Overview

A product design prompt about building a dashboard that tracks the quality of WhatsApp calls. It tests choosing the dashboard's audience, defining precise call-quality measures, picking breakdowns and alerting that localize problems, and prioritizing what ships in a first version.

WhatsApp call-quality dashboard: who it serves, what it measures, how it is laid out

Company: Meta

Role: Product Manager

Category: Product Design & Strategy

Difficulty: medium

Interview Round: Other

You are a product manager asked to design a dashboard that shows how good WhatsApp calls are. Decide who the dashboard serves and which decisions it should help them make, define the measures of call quality it reports, describe the views, breakdowns and any alerting it offers, and explain what belongs in the first version and what you would leave for later. ```hint Start from the viewer's first minute Pick the person who opens this dashboard and write down the question they need answered within a minute of opening it. Let that decide what sits at the top of the screen and what lives one click down. ``` ### Clarifying Questions - Who opens the dashboard: an on-call team watching for outages, the calling product team tracking trends and launches, or leadership checking overall health? - Does call quality here cover one-to-one voice calls only, or video and group calls as well? - Should quality be judged from technical signals the app collects during calls, from what people report after a call, or from both? - How fresh does the data need to be: within minutes for incident response, or daily for tracking trends? ### What a Strong Answer Covers - A chosen primary audience, the decisions the dashboard must support for them, and the scope settled up front (call types, data freshness) - A definition of call quality that spans what can go wrong before, during and at the end of a call, with each measure precisely defined - Breakdowns that let a viewer localize a problem, and a way to tell normal fluctuation from a real regression - A prioritized first version versus later additions, with the reasoning - How the dashboard's own success is judged, and the data and privacy risks that could make it mislead ### Follow-up Questions - The drop rate jumps in one country the day after a new app version starts rolling out; walk through how the dashboard helps you confirm and localize the cause. - How would you tell whether a falling post-call rating reflects worse calls or a change in who is asked to rate? - If leadership wants a single call-quality score, how would you build it and what would it hide?

Overview: A product design prompt about building a dashboard that tracks the quality of WhatsApp calls. It tests choosing the dashboard's audience, defining precise call-quality measures, picking breakdowns and alerting that localize problems, and prioritizing what ships in a first version.

Meta logo
Meta
Mar 14, 2024
mediumProduct ManagerOtherProduct Design & Strategy
0
0

You are a product manager asked to design a dashboard that shows how good WhatsApp calls are. Decide who the dashboard serves and which decisions it should help them make, define the measures of call quality it reports, describe the views, breakdowns and any alerting it offers, and explain what belongs in the first version and what you would leave for later.

Clarifying Questions Guidance

  • Who opens the dashboard: an on-call team watching for outages, the calling product team tracking trends and launches, or leadership checking overall health?
  • Does call quality here cover one-to-one voice calls only, or video and group calls as well?
  • Should quality be judged from technical signals the app collects during calls, from what people report after a call, or from both?
  • How fresh does the data need to be: within minutes for incident response, or daily for tracking trends?

What a Strong Answer Covers Guidance

  • A chosen primary audience, the decisions the dashboard must support for them, and the scope settled up front (call types, data freshness)
  • A definition of call quality that spans what can go wrong before, during and at the end of a call, with each measure precisely defined
  • Breakdowns that let a viewer localize a problem, and a way to tell normal fluctuation from a real regression
  • A prioritized first version versus later additions, with the reasoning
  • How the dashboard's own success is judged, and the data and privacy risks that could make it mislead

Follow-up Questions Guidance

  • The drop rate jumps in one country the day after a new app version starts rolling out; walk through how the dashboard helps you confirm and localize the cause.
  • How would you tell whether a falling post-call rating reflects worse calls or a change in who is asked to rate?
  • If leadership wants a single call-quality score, how would you build it and what would it hide?