Market Sizing: Daily Active WhatsApp Conversations Across India
Company: Meta
Role: Product Manager
Category: Product / Decision Making
Difficulty: medium
Interview Round: Other
Estimate how many WhatsApp chats take place across India on a typical day. State your assumptions, show the arithmetic, and explain how you would check that the answer is reasonable.
```hint Pin down the unit
Decide what counts as one chat before estimating anything, and remember that a single conversation involves at least two users, so per-user figures need care when you add them up.
```
### Clarifying Questions
- Does a "chat" mean a conversation thread that is active that day, or each individual message sent?
- Do group chats count, and should a group conversation count once or once per member?
- Are business accounts, broadcast messages, voice and video calls, and status updates in scope?
- Does "in India" mean users with Indian phone numbers, or only conversations where every participant is in India?
### What a Strong Answer Covers
- An explicit definition of a chat and a statement of what is in and out of scope.
- A funnel from India's population to daily WhatsApp users, with every ratio stated as an assumption.
- Per-user activity segmented by type of user and split between one-to-one and group conversations.
- Correct handling of shared conversations so they are not counted once per participant.
- A sanity check from an independent angle, and the assumptions that drive most of the uncertainty.
### Follow-up Questions
- How would your estimate change on a major festival day such as Diwali?
- If the interviewer meant messages rather than conversations, how would you convert your answer?
- Which single assumption moves your result the most, and how would you narrow it with public data?
Overview: A Meta product manager estimation question about WhatsApp: size how many chats take place across India on a typical day. It tests defining an ambiguous unit, building a funnel from population to daily users, segmenting activity by user type, avoiding double counting of shared conversations, and sanity-checking the final number.