Every major cloud provider's archive tier leads with a headline rate that looks almost too good to be true — a fraction of a cent per gigabyte, per month. On paper, that makes cold cloud storage look like an easy win over tape or on-prem disk. In practice, the number on the pricing page is rarely the number that ends up on your invoice. Four line items typically make the difference, and none of them show up in that headline rate.
1. Retrieval fees
Archive tiers are priced to store data cheaply on the assumption you'll rarely touch it. Pull that data back, and most providers charge separately for the privilege — sometimes per gigabyte retrieved, sometimes per request, sometimes both. For an archive that's genuinely cold, this barely matters. For a media archive with periodic restores — a rights renewal, a remaster, a compliance request — retrieval costs can rival or exceed the storage cost itself over the life of the content.
2. Minimum storage duration
Archive and cold tiers typically come with a minimum storage commitment — commonly 90, 180, or even 365 days, depending on the provider and tier. Delete or move an object before that window is up, and you're billed for the remainder of the minimum anyway. This matters more than it sounds: any workflow that migrates content between tiers, re-organizes an archive, or responds to a rights change earlier than expected can trigger early-deletion charges that quietly erode the savings that made the tier attractive in the first place.
3. Egress
Storing data in the cloud is usually cheap. Moving it back out — to the public internet, to another region, or in some cases even to another service within the same provider — is where cloud economics get less flattering. Egress is billed separately from storage and retrieval, and it's the cost most often left out of a back-of-envelope cloud migration estimate, right up until the first large restore or a provider switch makes it impossible to ignore.
4. Per-request and API charges
Object storage bills for operations, not just capacity — PUT, GET, LIST, and lifecycle-transition requests all carry a small per-request charge. For a handful of large media files, this is noise. For an archive with millions of small objects (proxy files, project metadata, individual frames), the per-request line item can add up to a meaningful chunk of the monthly bill, independent of how much data is actually stored.
What this means in practice
None of this makes cloud archive storage a bad choice — for a lot of media archives, it's still the most economical option once every real cost is accounted for. The point is that the headline $/GB/month figure is only one input among several, and the ones that get left out tend to be the ones that matter most for how an archive actually gets used: how often it's touched, how it's organized, and how often that organization changes. Modeling the full picture — storage, retrieval, minimum-duration exposure, and egress together — is what separates a realistic cost comparison from a pricing-page estimate.
Our storage cost calculator builds retrieval costs into its cloud comparison for this reason. If you want to work through what your own archive's real numbers look like, get in touch.
← Back to all posts