Quick Tip – Cut S3 Costs with Intelligent-Tiering

S3 storage costs have gotten complicated with all the storage classes and lifecycle options flying around. As someone who’s helped organizations cut their S3 bills significantly, I learned everything there is to know about quick wins that don’t require code changes. Today, I will share one of the best ones with you.

Here’s a quick tip for reducing S3 storage costs without touching your application code.

Enable Intelligent-Tiering

Probably should have led with this section, honestly. S3 Intelligent-Tiering automatically moves objects between access tiers based on usage patterns. Objects not accessed for 30 days move to infrequent access storage. No action required on your part.

After 90 days without access, objects move to archive tiers. Access patterns that resume move objects back automatically. That’s what makes this approach low-risk—it adapts to your actual usage without manual intervention.

Cost Calculation

Standard storage costs $0.023/GB in US East. Intelligent-Tiering costs the same for frequently accessed data, plus a small monitoring fee of $0.0025 per 1,000 objects.

Infrequent access tier costs $0.0125/GB—nearly 50% savings for dormant data. For buckets with a mix of hot and cold data, the savings add up quickly.

Implementation

Enable via bucket properties in the S3 console. New uploads automatically use the configured storage class. Takes about two minutes to set up.

Existing objects require lifecycle rules or batch operations to convert. Worth the effort for large buckets.

When to Skip This

Buckets with consistently accessed data see no benefit. The monitoring fee adds cost without savings if everything stays in the frequent access tier.

Very small objects (under 128KB) should use Standard storage due to minimum size charges. The monitoring overhead isn’t worth it for tiny files. For everything else though, Intelligent-Tiering is usually the right choice.