Storage Growth Projection
2 min read
Extrapolate how much storage you'll need in 6 and 12 months — not just day one.
Extrapolate how much storage you'll need in 6 and 12 months — not just day one.
How It Works
Storage estimates that only cover "today" miss the point. Databases, caches, and object stores become painful to migrate once they're full, so your estimate has to project forward. The formula: daily new data (users × actions per user × bytes per action) × retention period × a growth factor for user growth. For a URL shortener writing 10 million new URLs per day at 500 bytes each, that's 5 GB per day, or 1.8 TB per year at a flat user base. Factor in 10% month-over-month user growth and the year-one total is closer to 3 TB. The lesson: small daily numbers compound into large annual numbers fast. In interviews, always project at least 12 months ahead and state the growth assumption explicitly.
Real-World Example
Dropbox moved off AWS S3 in 2016 with a project called Magic Pocket, largely because projected storage costs over five years would have exceeded the capital cost of building their own datacenter. Storage-growth projection was central to that decision. The lesson: under-projecting storage doesn't just cost money at scale — it forces an expensive migration exactly when you can least afford one.
Test Yourself
Scenario: Your team runs a collaborative document editor. 500K DAU, each user creates an average of 3 documents per day, each document is ~50KB at rest. Plan year-one storage with 15% month-over-month user growth.
Get notified when we launch
One email when the full practice product is live. No spam.