All cheat sheets
Storagecheat sheet

S3 Inventory

A scheduled audit report of every object in a bucket — size, age, storage class, encryption status — delivered as queryable CSV/Parquet for ~$0.0025 per 1,000 objects. The foundation for cleanups, compliance answers, and Batch Operations manifests.

Last reviewed: July 11, 2026

Prefer step by step? Take the guided S3 Inventory walkthrough — free, anonymous, interactive.Have an AI assistant? Jump to the copy-paste audit prompt at the bottom of this sheet.

TL;DR: S3 Inventory answers "what is actually in this bucket?" — a scheduled (daily/weekly) report of every object with size, last-modified, storage class, encryption status, and more, delivered as CSV or Parquet you can query with Athena. At $0.0025 per 1,000 objects it costs pennies, and it's the manifest source for Batch Operations. You can't optimize what you can't list.

The numbers

  • Cost: $0.0025 per 1,000 objects listed — a million-object bucket reports for $2.50
  • Field-tested finds from the source workflow: 12,000 orphaned profile pictures (objects with no DB row); 400+ duplicate log files found by matching ETags; an inherited 15-bucket estate cut from $140/month to $22 (~$1,400/year) after one inventory pass showed 11 buckets were dead weight
  • Compliance answer in a report: "97% encrypted, 3% not" — with the exact keys to fix

Do this

  1. Configure it on your biggest (or most mysterious) buckets: console → bucket → Management → Inventory configurations, or put-bucket-inventory-configuration. Choose Parquet if there's any chance you'll query it; include size, last-modified, storage class, encryption status, and (on versioned buckets) current-versions-only.

  2. Wait for the first report (up to 48 h daily / 7 days weekly), then point Athena at the destination and ask the questions that have been bothering you:

    SELECT key, size, last_modified_date FROM s3_inventory
WHERE size > 104857600
  AND last_modified_date < date_add('year', -1, now())
ORDER BY size DESC;

    Instant cleanup hit list: big, old, forgotten.

  3. Cross-reference against your app database for orphan detection — S3 keys with no corresponding row are pure waste.

  4. Feed the results into action: Batch Operations consumes inventory manifests directly for the one-time cleanup; lifecycle policies prevent re-accumulation. The mature loop is inventory → Athena filter → batch job → Slack report.

Gotchas

  • Don't point the destination inside the source scope — the next report inventories the previous reports, recursively.
  • Keep the destination in the same region (cross-region delivery bills transfer fees) and lock it down — inventory metadata (filenames, encryption flags) is sensitive.
  • Versioned buckets bloat reports unless you select current-versions-only.
  • Expire old reports with a lifecycle rule on the destination (90 days is plenty) — hoarding weekly reports from 2022 is its own little waste stream.
  • Reports are snapshots, not real-time — eventually consistent, on a schedule. Event-driven work belongs to S3 Notifications.

Skip this if

  • The bucket is small enough to eyeball in the console.
  • You only need aggregate metrics (total size, class distribution) — that's Storage Lens, no per-object listing needed. Lens spots the what; Inventory gets you the list.

Run this audit with your AI assistant

Paste this into Claude, ChatGPT, or any agent that can run the AWS CLI with read-only credentials. It audits your account for exactly the waste this sheet describes — and changes nothing.

You are auditing an AWS account's use of S3 Inventory and identifying
where it should be enabled. Use the AWS CLI with READ-ONLY credentials.
Do not create, modify, or delete anything — report findings and
recommended (unapplied) fixes only.

1. Which buckets have Inventory configured?
   aws s3api list-bucket-inventory-configurations --bucket <b>
   For configured ones: note fields captured, format (CSV vs Parquet),
   schedule, destination — and whether destination has a lifecycle rule
   expiring old reports (report hoarding is its own waste).
2. Which large buckets lack Inventory? Rank by CloudWatch
   BucketSizeBytes; recommend weekly Parquet inventory for the top
   buckets, prefix-filtered where only part of a bucket matters.
3. Sanity checks on configs: destination bucket must not be inside the
   source scope (recursive reports); destination should be same-region
   (cross-region delivery costs transfer fees); versioned buckets
   should usually report current versions only.
4. Cost estimate: $0.0025 per 1,000 objects listed per run.

Report: existing configs with issues flagged; recommended new configs
as put-bucket-inventory-configuration JSON ready for review — but do
NOT apply them; plus 3 example Athena queries the user should run on
the first report (oversized+stale objects, unencrypted objects,
storage-class distribution).
Works with any assistant that can run shell commands.

Want the guided version?

The S3 Inventory walkthrough covers this topic interactively — it asks about your setup, branches to what’s relevant, and quizzes you on the tricky parts. Free and anonymous.

Start the walkthrough

More Storage cheat sheets

AWS Backup Archive Tier

One lifecycle field on a backup plan moves long-retention recovery points to Glacier-backed cold storage at ~$1/TB-month — an 80–90% cut on backups whose realistic future is 'sits on a shelf until the retention clock runs out.'

EBS DeleteOnTermination

The wrong default on secondary EBS volumes leaves orphaned storage billing forever after instances terminate. One boolean in your launch templates stops the bleed; one audit command finds what already leaked.

EBS gp3 Volumes

Migrate EBS volumes from gp2 to gp3 for an immediate 20% price cut with the same or better performance — live, no downtime, one API call per volume.

EBS Snapshot Archive

Move compliance backups and 'just in case' EBS snapshots from Standard to the Archive tier for 75% lower storage cost — the trade is a 24–72 hour restore window that cold snapshots never actually need.