← Back to all posts

Cloud Storage for Business: What It Means and Where to Start

3 September 2026 · 3 min read

“Move it to the cloud” is advice given far more often than it is explained. For a business deciding where its documents, records and backups should live, the useful starting point is that cloud storage is not one product — it is three quite different ones, and mixing them up is where money gets wasted.

The three kinds, and what each is for

File sync and share — Google Drive, OneDrive, Dropbox. Built for people. Folders, sharing links, editing in the browser, a familiar interface. This is where your team’s working documents belong.

Object storage — Amazon S3, Azure Blob, Google Cloud Storage and similar. Built for applications. You address files by name through an API rather than browsing folders. This is where application data, media libraries, archives and backups belong. It is dramatically cheaper at volume than the consumer-facing services, and considerably less pleasant to use by hand.

Block storage — the virtual hard disk attached to a server. Only relevant if you run servers.

Most small businesses need the first, quietly benefit from the second behind the scenes, and never touch the third directly. Paying sync-and-share prices to archive a decade of files is one of the more common avoidable costs we see.

What actually drives the bill

Storage itself is cheap and keeps getting cheaper. The costs that surprise people are the other three:

  • Egress — what you pay to get data out. Storing is cheap; bulk retrieval can be expensive. This is also what makes leaving a provider costly, so check it before you commit rather than after.
  • Requests — a charge per operation. Irrelevant for documents, material for applications making millions of small reads.
  • Storage class — archive tiers cost a fraction of standard storage but may take hours to retrieve. Excellent for records you must keep and hope never to need; useless for anything operational.

The GDPR questions to settle first

If you hold personal data — customers, employees, applicants — these are not optional:

  1. Where is it physically stored? Most major providers let you pin data to an EU region. Choose deliberately; the default may not be Europe.
  2. Is there a data processing agreement? Your provider is a processor and you need this in place. Business tiers include one; free consumer accounts generally do not, which is a real problem if staff are using personal accounts for work.
  3. Who can access it? Least privilege, and remove access when people leave. Dormant accounts with live permissions are a recurring finding.
  4. How long do you keep it? GDPR expects a retention period, not “forever”. Object storage lifecycle rules can enforce this automatically.

This overlaps directly with your AI tool policy — the same questions apply when a vendor processes your documents.

A sensible starting structure

Working documents in a business-tier sync-and-share account with proper access control. Archives and backups in object storage with a lifecycle policy. Personal data pinned to an EU region under a signed processing agreement. Access reviewed when anyone joins or leaves.

That covers the great majority of small-business needs without a migration project.

One thing cloud storage is not

Cloud storage is not a backup. Sync is not backup — delete a file, or encrypt it with ransomware, and the change propagates to every synced copy within seconds. That distinction deserves its own article, and has one: backups are not a cloud feature.

Not sure whether your current setup is sensible or expensive? Get in touch — a short review usually pays for itself.