Estimate infrastructure or operational capacity needed for Slack at global scale

Problem Statement Description

Product context: Salesforce is an enterprise CRM and cloud software company; its products include Sales Cloud, Service Cloud, Marketing Cloud, Commerce Cloud, Data Cloud, Einstein AI, Tableau, and Slack. Slack is Salesforce's workplace collaboration product; its products include channels, DMs, huddles, workflow automation, search, app integrations, and enterprise administration.

Estimate the infrastructure or operational capacity Slack would need to reliably support global-scale usage in an enterprise environment, with Salesforce as the company context. The focus is on Slack as a collaboration layer for large organizations, including sales leaders and revenue teams who rely on real-time messaging, channels, workflow automation, CRM integrations, notifications, search, file sharing, and AI-enabled productivity features.

You should define the scope of the estimate clearly: what “global scale” means, which user populations are included, what usage behaviors matter, and which capacity units are most relevant. The estimate may include technical infrastructure capacity such as concurrent users, messages per day, storage, API calls, search indexing, notifications, compute, and network traffic, as well as operational capacity such as reliability coverage, customer support, Trust & Safety, compliance, or incident response.

This is a guesstimate question, so the emphasis is on structuring the problem, making transparent assumptions, sizing the major drivers, and pressure-testing the result. You are not expected to know Slack’s internal numbers, but your approach should reflect an understanding of enterprise SaaS usage, collaboration workflows, and the reliability expectations of global Salesforce customers.

The experience should consider:

- The target population, including enterprise employees, sales teams, external collaborators, admins, apps, and bots.

- Adoption assumptions across regions, company sizes, industries, and paid enterprise customers.

- Usage frequency assumptions such as daily active users, concurrent users, messages per user, channel activity, file uploads, workflow runs, and notifications.

- Key capacity units, including compute, storage, bandwidth, database reads/writes, search indexing, API throughput, and peak-load handling.

- Differences between average usage, peak business-hour usage, regional spikes, and incident or event-driven surges.

- Operational requirements such as uptime, latency, compliance, data residency, customer support, incident management, and security monitoring.

- Sensitivity checks around high-growth scenarios, AI feature adoption, heavy CRM integration usage, and large enterprise deployments.

- Sanity checks comparing the final estimate against plausible enterprise SaaS benchmarks and user behavior patterns.

The goal is to produce a reasoned, defensible capacity estimate that shows how you break down an ambiguous global-scale infrastructure problem, identify the highest-impact drivers, and communicate assumptions and uncertainty clearly.