Opens in a new tab.

Our approach: Assess, find the gaps; Fix, put it right; Manage, keep it working; Assure, prove it stays that way.

← All insights DIGIPEST / INSIGHT

A practical cyber security baseline for small organisations

The essential controls small organisations can put in place to reduce common cyber risks without building an enterprise security programme.

4 min readBy DigiPest

Cyber security can sound like an open-ended commitment: more products, more monitoring and more rules. For a small organisation, that is not a useful starting point.

A better question is: which failures would hurt the organisation most, and which basic controls make those failures substantially less likely?

The following baseline is not a guarantee against every attack. It is a practical way to reduce the common risks that cause disproportionate disruption to small businesses, charities and other organisations with limited internal IT resources.

Know what you depend on

Start with a simple record of the technology the organisation needs to operate. Include:

  • laptops, desktops, phones and tablets;
  • Microsoft 365, Google Workspace and other cloud services;
  • websites, domains and email systems;
  • routers, firewalls, wireless access points and remote-access services;
  • important suppliers and externally managed systems; and
  • the business information each system stores or processes.

This does not need to become a complicated asset-management project. It needs to be accurate enough to identify unsupported equipment, forgotten accounts and systems that nobody clearly owns.

Keep supported systems updated

Security updates fix known weaknesses. Once a weakness and its fix are public, attackers can look for organisations that have not applied it.

Enable automatic updates where that is appropriate, establish who checks the systems that cannot update automatically, and replace products that no longer receive security fixes. Include browsers, mobile devices, network equipment and third-party applications—not only Windows updates.

Updates should also be managed sensibly. Business-critical changes may need testing, but indefinite delay is not a testing strategy. Record exceptions and give them a review date.

Protect accounts, especially email and administration

Use a unique account for each person, remove access promptly when someone leaves, and avoid using administrator accounts for routine work.

Enable multi-factor authentication or passkeys wherever they are supported, prioritising email, cloud administration, finance, backups and remote access. Use a managed password manager rather than expecting people to invent and remember a different complex password for every service.

Back up the information needed for recovery

Decide what must be recoverable, how quickly it is needed and how much recent work the organisation could afford to lose.

Backups should not rely on the same account, device or cloud service as the live data. Ransomware and account compromise can affect synchronised folders and reachable backups. Keep an appropriately isolated copy, restrict administrative access and test restoration—not merely whether a backup job reports success.

Configure devices and networks deliberately

Change default credentials, enable built-in firewalls and malware protection, encrypt portable devices, and require screen locks. Remove services and accounts that are no longer needed.

Review routers, firewalls and wireless networks as business systems. They need supported firmware, controlled administration and a clear record of who can change their configuration.

Help people recognise and report problems

People should know how to question an unusual payment request, report a suspicious message and obtain help after clicking something they should not have clicked.

The reporting route matters as much as awareness training. A quick report can turn a compromised password into a manageable incident. Fear of blame encourages delay.

Prepare for the day a control fails

Write down who will make decisions, who will contact the IT provider, which services should be isolated, and how customers or regulators would be informed if necessary. Keep essential contact details somewhere accessible when normal systems are unavailable.

An incident plan can be short. Its value comes from being usable under pressure and tested before the organisation needs it.

Turn the baseline into evidence

Assign an owner and review date to each control. Keep evidence such as device inventories, update reports, backup tests, access reviews and incident exercises. That makes it possible to distinguish a control that exists from one that is merely assumed to exist.

The NCSC small-organisation guidance provides a useful starting point. Organisations seeking a recognised baseline can also work towards Cyber Essentials .

The objective is not perfect security. It is a known, maintained set of protections that makes common attacks harder and recovery more realistic.