Meta PM Interview — WhatsApp India Engagement Metrics

It tests whether you can define what good looks like before proposing anything. You are expected to pick one north-star metric tied to real user value, give it a denominator, decompose it into input metrics the team can move, name counter-metrics and guardrails that protect quality and trust, and state a measurable target with a timeframe. For a WhatsApp India prompt it also tests whether you segment users by tier-1, tier-2, and tier-3 and reason about data cost and low-end devices. The prompt is open-ended on purpose, you define metrics from scratch rather than picking from a list.

Start by clarifying the goal and the specific user segment you are trying to move, not a feature. State one north-star with an explicit numerator and denominator so it cannot be gamed. Break it into a small number of input metrics that actually drive it. Name at least one counter-metric or guardrail that protects spam, quality, or trust. Then commit to a measurable target and a timeframe and explain how you would measure it, including experiment design. Only after the metric spine is in place should you discuss tactics. Keep the India context concrete throughout.

The biggest mistake is naming a feature or tactic before defining the goal and the segment. The second is picking total messages sent as the headline number with no denominator, which is a vanity metric. Others include never naming a counter-metric, stating a north-star but never giving a target or timeframe, jumping between metrics with no decomposition, and ignoring India realities like data cost, low-end Android, and tier-2 and tier-3 behaviour. Candidates also lose points when they quote a metric with no measurement method or attribution reasoning.

The behaviour is modelled on a real analytical-thinking loop interviewer. It asks one question at a time, probes every answer at least once, and never praises you mid-round. The difference is consistency and recall. It will not get distracted, it holds you to the same depth on every claim, and it produces a transcript-backed scorecard afterward that quotes the exact moments your metric reasoning held or broke. A human interviewer may vary; this keeps the bar identical every run so you can practice deliberately.

Your answer is evaluated against observable behaviours, not delivery style. The scorecard looks at whether you defined a single north-star with a denominator, whether you decomposed it into movable input metrics, whether you named counter-metrics and guardrails, whether you set a measurable target and timeframe, and whether you segmented Indian users before proposing tactics. Each dimension is scored from the transcript with examples, so two evaluators would land close. Ideas and structure are scored, not accent or fluency.

Do not pitch a feature. Spend the opening restating the goal in your own words and naming exactly which Indian users you are trying to move and why they are or are not engaging today. Ask one or two sharp clarifying questions if scope is genuinely ambiguous. Then commit to a candidate north-star out loud and say what its denominator is. Getting the goal and segment crisp in the first two minutes is the single strongest signal in this round and the thing most candidates skip.

Do not defend it reflexively. Acknowledge the specific gap, usually a missing denominator or a quality blind spot, and reframe the metric on the spot. For WhatsApp India that often means moving from total messages to something per active user with a quality qualifier, like two-way conversations per weekly active user. Then state what counter-metric protects it. Showing you can recalibrate a metric under pressure without abandoning the goal is exactly what this round rewards.

A strong answer sounds like: here is the goal, here is the user segment and why they behave the way they do, here is my one north-star with its numerator and denominator, here are the two or three input metrics that move it, here is the counter-metric that keeps me honest on spam and trust, here is the target and the timeframe, and here is how I would measure and attribute it. It stays concrete about tier-2 and tier-3 India and it never leans on total messages without normalising it.

India is WhatsApp's single largest market, with roughly 535 to 536 million monthly active users in 2025 and a path toward one billion by the end of 2026. The scale, the tier-1 to tier-3 spread, the data-cost and device constraints, and the growing Meta AI and commerce layers make it a rich, real goal-setting problem. It forces a candidate to choose a north-star under genuine ambiguity, which is exactly the analytical-thinking skill the Meta loop is built to test.

The practice round runs about twenty minutes across a warm-up, a core metric-definition block, a pressure block, and a short reflection. You walk away with a transcript-backed scorecard that names the specific dimensions where your metric reasoning held and where it broke, for example whether your north-star had a denominator or whether you ever named a counter-metric. It is built for deliberate repetition, not a single attempt.