Estimation exercise: total daily message volume on Slack
Company: Microsoft
Role: Product Manager
Category: Product / Decision Making
Difficulty: medium
Interview Round: Other
Estimate how many messages are sent on Slack in one day.
```hint Workdays matter
Slack is a work tool, so think about how a typical weekday differs from a weekend before you average, and decide whether posts from bots and app integrations belong in the count.
```
### Clarifying Questions
- Should the count cover every Slack workspace worldwide, free and paid, or only paying customers?
- Do automated posts from bots and app integrations count, or only messages written by people?
- Does a thread reply or a direct message count the same as a channel post, and do emoji reactions or edits count at all?
- Is "per day" an average over all seven days, or a typical weekday?
### What a Strong Answer Covers
- A clear definition of a sent message, settled before estimating
- A defensible figure for daily active Slack users, labeled as an assumption and cross-checked another way
- Segmentation of users by how much they message, with round numbers
- Adjustments for weekends and for automated messages
- A sanity check, and which assumption moves the answer the most
### Follow-up Questions
- How would you estimate how many messages are read per day, given that one channel post reaches many people?
- If you were comparing Slack with Microsoft Teams, which matters more from this estimate, total messages or messages per active user, and why?
- How would you turn this into an estimate of the data Slack has to store each day?
Overview: A product manager estimation exercise asking how many messages are sent on Slack each day. It tests defining what counts as a message, estimating daily active users with a cross-check, segmenting users by activity, adjusting for weekends and bots, and sanity-checking the result.