HardExperience level: 8-15 years1 min read

How do you do back-of-the-envelope capacity estimation in a design interview?

Reviewed by Gurusankar M.

Asked inAmazonGoogleMicrosoftMeta-style
#capacity estimation#back of envelope#qps#storage#bandwidth
Report issue

Short Answer

Derive QPS from DAU × actions/day ÷ 86,400 (and apply a peak factor of ~2–5×). Estimate storage from objects/day × size × retention, and bandwidth from QPS × payload size. Round to powers of ten — the goal is order-of-magnitude sizing to justify architecture (caches, shards, replicas), not precision.

On this page

Concept Made Simple

Coffee Chat Question

“How do you do back-of-the-envelope capacity estimation in a design interview?”

Remember It Faster

Mind Map Answer

QPS
DAU × actions ÷ 86,400 × peak factor
Storage
items/day × size × retention
Bandwidth
QPS × payload size
Goal
order-of-magnitude, not exact

Learn by Doing

Hands-on Keyboard

text
100M DAU, 10 reads/user/day:
  reads/day = 1e9
  avg QPS   = 1e9 / 86,400 ≈ 11.6k
  peak QPS  ≈ 11.6k × 3 ≈ 35k
2 KB/response → peak BW ≈ 35k × 2KB ≈ 70 MB/s

Think Beyond the Expected

What If?

Why bother estimating if the numbers are rough?

The magnitude drives the architecture: 35k QPS says you need caching, multiple app instances, and read replicas/sharding — a single DB won't do. Estimation justifies WHY you add each component and shows the interviewer you can size a system, which is the point (not the exact number).

Real World

Every senior system-design interview expects quick capacity math to justify caches, shards, and replicas; the skill is reasonable assumptions + powers-of-ten arithmetic, not precision.

Interviewer's Expectation

Keywords they're listening for:

  • QPS from DAU × actions
  • peak factor
  • storage = rate × size × retention
  • bandwidth = QPS × payload
  • order-of-magnitude, justify architecture

Common Mistakes

  • Skipping estimation entirely
  • Forgetting the peak factor
  • Over-precise math instead of magnitudes

Best Practices

  • State assumptions; round to powers of ten
  • Apply a peak/burst multiplier
  • Tie the numbers to architecture choices

Follow-up Questions

  1. How does the estimate justify adding a cache or shards?
  2. Read:write ratio — why does it matter?
  3. How do you estimate storage growth over a year?

Continue Learning with AI

Take this question deeper with your favourite AI assistant. Pick a depth, copy the prompt, or open it directly — AI is your learning companion, not a shortcut.

Plain-language foundations

I'm preparing for a software engineering interview and want to understand this from scratch, as a beginner.

Topic: Scaling (System Design)
Interview question: "How do you do back-of-the-envelope capacity estimation in a design interview?"

Please:
1. Explain the core idea in simple, plain language, using an everyday analogy.
2. Define any technical terms you use.
3. Walk through one small, concrete example.
4. Finish with a single sentence I can easily remember.

Keep the tone friendly and assume I'm new to this topic.
Open inChatGPTGeminiClaude

Was this answer helpful?

Affiliate links

Tools for Your Developer Workspace

Equipment for a comfortable, productive setup.

As an Amazon Associate I earn from qualifying purchases.