Azure blob storage offers hot, cool, cold, and archive access tiers, and archive is by far the cheapest to store but the most expensive and slowest to read. Archived blobs are offline: you must rehydrate them to an online tier first, which takes hours and carries a retrieval charge that scales with how fast you want them back. The economics are simple to state and easy to get wrong. If data is large, must be kept for a long time, and is read rarely with no urgency, archive wins decisively. If you might read it on short notice, or it is young, or it is millions of tiny blobs, archive can cost more than the tier you came from.
This sits in the Azure storage cluster with blob tiers and lifecycle management and Azure storage cost optimization end to end. Read those for the full tier ladder and the lifecycle automation that moves data into archive safely.
How cheap is the archive tier really?
On storage price per gigabyte, archive is dramatically lower than hot or cool, which is what makes it tempting for compliance retention, old backups, and raw data you are required to keep. But storage price is only one of three cost components, and for cold data the other two are what decide the outcome.
The components are storage (very low in archive), access and transaction costs (higher than the warm tiers), and rehydration retrieval (a charge to bring data back online, larger for high priority). For data you genuinely never read, only storage matters and archive is unbeatable. For data you read even occasionally, the access and rehydration costs start eating the storage saving, and the break even depends on how often and how urgently you read.
What is rehydration, and what does it cost?
This is the part buyers miss. An archived blob cannot be read in place. To access it you rehydrate it to the hot or cool tier, a process that takes hours, with a standard priority that is cheaper and slower and a high priority that is faster and dearer. You pay a retrieval charge for the rehydrated data, and you then pay the online tier storage price for as long as it stays rehydrated.
The trap is treating archive as if it were just a cheaper cool tier. A team that archives data and then finds it needs to read a meaningful fraction of it each month pays repeated rehydration and retrieval charges that can exceed what cool storage would have cost outright, plus they wait hours for data they expected instantly. Archive is a one way bet that the data stays cold; price the cost of being wrong.
When should you not use archive?
Three clear no go cases, plus the early deletion rule.
- Latency sensitive reads. If anyone might need the data within minutes, archive's multi hour rehydration disqualifies it regardless of price.
- Frequent reads. If you read the data more than rarely, accumulated access and rehydration costs beat the storage saving. Cool or cold is the right floor for warm but infrequent data.
- Many tiny blobs. Per object and per transaction overhead dominates at high object counts, so aggregate first or keep them in a warmer tier.
There is also an early deletion charge: archive has a minimum retention period, and deleting or moving a blob before it elapses incurs a prorated early deletion fee. Do not archive data with a short or uncertain lifespan, because you may pay to delete it.
A worked example
A European SaaS company held an indicative petabyte scale pool of raw event data and compliance records. They split it correctly. Records under a multi year legal hold, large, never read except in a rare audit, went to archive via a lifecycle rule once they passed the early deletion threshold, and the storage line for that pool fell by a large multiple against cool. A second pool, recent event data that analysts occasionally re queried, was kept in the cold tier rather than archive, because modelling showed the rehydration churn from even monthly reads would have wiped out the storage saving and added hours of latency. Same data type, two different tiers, each chosen on read pattern, not on the headline storage price. Figures are verified against billing data and anonymised.
Frequently asked questions
How cheap is the Azure archive tier?
What is rehydration and what does it cost?
When should you not use the archive tier?
Where archive fits the storage review
Archive is one rung of a blob tier ladder that a lifecycle policy should manage automatically, and getting the read pattern right is the whole game. We run that storage review as an independent buyer side advisory across AWS, Azure, GCP, and OCI, with zero provider commissions and a guarantee: we reduce your cloud spend or we reimburse our service fee.
For monthly buyer side storage analysis, subscribe to The Cloud Spend Navigator.
Put a defensible number on your cloud spend.
No provider in the room, no published price list. Tell us your footprint and we will scope the savings against your billing data — we reduce your cloud spend or we reimburse our service fee.
The Cloud Spend Navigator: what changed in cloud pricing, commitments, and FinOps — no vendor spin.