📈 Storage Breakpoints: S3 vs On-Prem (with numbers)

Assumptions (conservative, operator-realistic):


< 300 TB

MetricS3On-prem
Annual storage cost~$10k~$5k
Infra headcount00.25–0.5 FTE
People cost$0$50k+
Net economicsS3 winsLoses

Why:
You’re optimizing the wrong line item. One infra engineer costs more than the entire storage bill.

Operator note:
If you move off S3 here, it’s ideology, not economics.


300–500 TB

MetricS3On-prem
Annual storage cost$15k–18k$9k
Total effective cost~$20k~$60k
Savings

Why:
You can run on-prem, but it’s a rounding error compared to staffing + risk.

Signal:
If storage outages won’t kill your business, you’re still too small.


~1 PB (the gray zone)

MetricS3On-prem
Annual storage~$35k~$18k
Egress (10%)~$110k~$5k
People + ops$0~$120k
Total~$145k~$143k

Why it “depends”:

Operator truth:
This is where finance says “maybe” and infra says “are you sure?”


🔸 3–5 PB (clear crossover)

Example: 5 PB

MetricS3On-prem
Storage~$175k~$90k
Egress~$500k~$25k
Infra team$0~$250k
Annual total~$675k~$365k
Delta~$310k saved/year

Why on-prem wins:

This is the “Jeremy zone.”


🔹 10+ PB (no contest)

Example: 10 PB

MetricS3On-prem
Storage~$350k~$160k
Egress~$1M+~$50k
Infra team$0~$350k
Annual total~$1.35M+~$560k
Savings$750k–900k/year

Why S3 breaks here:

On-prem at this scale becomes a moat.


🧠 Mental model to remember

S3 is O(n).
On-prem is O(1) + O(people).

Once people cost flattens, every extra PB massively favors on-prem.


⚠️ Final operator warning

On-prem only wins if:

Otherwise:

S3 is expensive — but organizational failure is worse.

If this sounds like your situation, let's talk

No pitch decks. No sales process. Just a conversation about what you're dealing with and whether I can help.

Book a 30-minute call →