Azure Blob Storage Tiers: Hot vs Cool vs Cold vs Archive — The 2025 Cost Decision Guide
Azure Blob Storage tiers differ not just in price per GB stored, but in access cost, retrieval time, and minimum retention requirements — choosing the wrong tier is a common and costly mistake
Azure cloud storage bills surprise many organisations because the headline price — cost per GB stored — is only one component of the total cost. Azure Blob Storage tiers differ significantly in their access costs, early deletion penalties, and retrieval times. An organisation storing 100TB of data in the Hot tier when Cool or Cold would suffice is typically overpaying by 40–60% on storage alone. An organisation that moved data to Archive to save money but needs to retrieve it regularly is incurring retrieval costs that may exceed what they would have paid to leave it in Hot.
This guide explains the four Azure Blob Storage tiers Hot, Cool, Cold, and Archive — with 2025 pricing data, a practical decision framework, and the lifecycle management approach that eliminates the need to manage tier selection manually at scale.
Why Azure Storage Tier Selection Is a CFO-Level Decision
For organisations with data estates measured in terabytes or petabytes, storage tier selection is not an infrastructure detail — it is a cost driver with direct P&L impact. A 100TB data estate stored entirely in Azure Blob Hot tier costs approximately ,000–,300 per month in storage alone (LRS, East US region, 2025 pricing). The same estate in Cool tier costs approximately ,000–,200 per month. In Cold tier, approximately 50–00 per month. In Archive, approximately 00–30 per month — but with retrieval costs and times that make it unsuitable for any data accessed regularly.
Most organisations have not performed a systematic analysis of which data actually needs to be in which tier. The result is a mixed estate where frequently accessed data is correctly in Hot, but months or years of historical data — accessed rarely or never — is also in Hot, generating unnecessary storage cost month after month. Implementing the right tier strategy and lifecycle policy typically reduces Azure storage spend by 30–60% without any change to data availability or analytical capability.
The Four Azure Blob Storage Tiers Explained
Hot tier
Hot is optimised for data that is accessed frequently — multiple times per day or per transaction. It has the highest storage cost per GB but the lowest access cost per operation. Use Hot for active operational data, current-period reporting datasets, and any data accessed in real time by applications or analytics pipelines. In Microsoft Fabric and Azure Data Lake Storage Gen2 contexts, active Delta tables and live Fabric Lakehouse data should be in Hot.
Cool tier
Cool is optimised for data accessed infrequently but that still needs to be available within milliseconds. There is a minimum 30-day retention requirement — data deleted or moved before 30 days incurs an early deletion charge. Use Cool for prior-quarter reporting data, completed project datasets, infrequently accessed backups, and historical data that might be needed for audit or investigation but is not queried regularly.
Cold tier
Cold (introduced in 2023) fills the gap between Cool and Archive. It has a 90-day minimum retention requirement, storage costs approximately 35% lower than Cool, and access costs higher than Cool but significantly lower than Archive retrieval. Cold is well-suited for compliance archives that must be accessible within minutes, data retained for regulatory requirements with occasional access, and historical datasets queried less than once per month.
Archive tier
Archive is the lowest storage cost option — approximately 95% cheaper than Hot per GB — but data in Archive is offline. Retrieval requires rehydration to Hot or Cool, which takes between one hour (high priority) and 15 hours (standard priority), and incurs a per-GB retrieval fee. Use Archive only for data that is retained for compliance or legal hold purposes and is genuinely never accessed in normal operations. The minimum retention requirement is 180 days.
Cost Comparison: Hot vs Cool vs Cold vs Archive in 2025
| Tier | Storage (per GB/month) | Read operations (per 10K) | Retrieval (per GB) | Min retention | Access latency |
|---|---|---|---|---|---|
| Hot | ~/bin/sh.018 | ~/bin/sh.004 | Free | None | Milliseconds |
| Cool | ~/bin/sh.01 | ~/bin/sh.01 | ~/bin/sh.01 | 30 days | Milliseconds |
| Cold | ~/bin/sh.0045 | ~/bin/sh.05 | ~/bin/sh.03 | 90 days | Milliseconds |
| Archive | ~/bin/sh.00099 | ~.50 | ~/bin/sh.02 | 180 days | 1–15 hours |
Indicative pricing for Azure East US region, LRS redundancy, June 2025. Verify current pricing at azure.microsoft.com/pricing/details/storage/blobs.
The Tier Selection Decision Framework
| Access frequency | Retrieval requirement | Recommended tier |
|---|---|---|
| Daily or multiple times daily | Milliseconds | Hot |
| Monthly or less | Milliseconds | Cool |
| Quarterly or less | Minutes acceptable | Cold |
| Rarely or never (compliance only) | Hours acceptable | Archive |
Azure Storage Lifecycle Management: Automate Tier Transitions
Managing tier selection manually at scale is impractical. Azure Storage Lifecycle Management allows organisations to define policy rules that automatically transition blobs between tiers based on age, last-access time, or custom conditions — without any manual intervention.
A typical enterprise policy might transition blobs to Cool after 30 days of no access, to Cold after 90 days, and to Archive after 365 days. This policy can be applied at the container or blob prefix level, allowing different data types to follow different retention schedules. Organisations that implement lifecycle policies typically see storage cost reductions of 40–60% within the first three months, as historical data that has been accumulating in Hot is automatically moved to appropriate lower-cost tiers.
For organisations running Microsoft Fabric or Azure Data Lake Storage Gen2, lifecycle policies apply at the storage account level and work seamlessly alongside Fabric's OneLake — though Fabric-managed data should be excluded from lifecycle policies to prevent unintended tier transitions that could affect query performance.
Five Costly Mistakes Organisations Make with Azure Storage Tiers
1. Defaulting everything to Hot. The most common and expensive mistake. When storage accounts are created, Hot is the default — and many organisations never revisit the tier assignment as data accumulates over months and years.
2. Moving frequently accessed data to Archive to chase the lowest storage rate, then incurring retrieval costs that eliminate the saving. Archive is only cost-effective for data accessed less than once per year.
3. Ignoring early deletion penalties. Moving data out of Cool before 30 days, Cold before 90 days, or Archive before 180 days incurs an early deletion charge equivalent to the remaining retention period. Rapid tier changes in response to unexpected access needs can generate substantial unexpected charges.
4. Not accounting for transaction costs when evaluating the total cost of lower tiers. Cool and Cold have meaningfully higher read operation costs than Hot. For data accessed frequently in small chunks (high read operation count), the transaction cost can eliminate the storage saving.
5. Applying a single policy to all data types. Raw ingestion data, processed analytical datasets, compliance archives, and application backups have very different access patterns and retention requirements. A single lifecycle policy applied uniformly will inevitably be suboptimal for some data types.
- Azure Blob Storage has four tiers — Hot, Cool, Cold, and Archive — with storage costs ranging from ~/bin/sh.018/GB (Hot) to ~/bin/sh.001/GB (Archive). Choosing the wrong tier can mean paying 18x more than necessary for infrequently accessed data.
- Hot = daily access. Cool = monthly access. Cold = quarterly access.Archive = compliance hold only — retrieval takes 1–15 hours.
- Minimum retention requirements (Cool: 30d, Cold: 90d, Archive: 180d) create early deletion penalties — plan tier transitions with retention periods in mind.
- Azure Storage Lifecycle Management automates tier transitions based on age and access patterns — implementing lifecycle policies typically reduces storage spend by 40–60% within three months.
- Exclude Microsoft Fabric and OneLake-managed data from lifecycle policies — automatic tier transitions can affect Fabric query performance on active Delta tables.
Next Steps: Azure Storage Cost Optimisation
A storage tier audit is one of the fastest ways to reduce Azure cloud costs without any change to your data architecture or analytical capability. Numlytics performs Azure data platform assessments that include storage tier analysis, lifecycle policy design, and cost modelling — typically identifying 30–60% storage cost reduction opportunities within the first week of engagement. Speak with an Azure data platform consultant to understand what a storage optimisation assessment would find in your current environment.