Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is Defence in Depth?

Defence in depth is a security strategy that stacks several independent, different controls around an asset, so that when one layer fails the next one slows the attacker down and makes them visible.

Illustration for the cybersecurity term Defence in Depth
Home / Blog / Cybersecurity Terms / What Is Defence in Depth?
17 November 2026

What Is Defence in Depth?

What Is Defence in Depth?

Defence in depth is the practice of protecting an asset with several layers of security controls that work in different ways, so that no single failure exposes the whole environment. It assumes every control will eventually be bypassed, misconfigured, or switched off, and plans for that moment by putting something else in the attacker’s path.

The idea comes from military doctrine: a castle relied on a moat, outer walls, an inner keep, and watchmen rather than one wall. The goal was never to make an assault impossible, but to make it slow and noisy enough for defenders to notice and respond. The same logic applies to networks, cloud tenancies, and applications.

Why One Strong Control Is Not Enough

Most serious incidents are not the result of one dramatic failure. They grow from a chain of small gaps that line up—a reused password, a flat network, an unmonitored server, an over-privileged account. Safety engineers call this the “Swiss cheese model”: each slice has holes, but a hazard only gets through when the holes align.

The Colonial Pipeline ransomware attack in May 2021 is a well-known example. Attackers reportedly entered through a legacy VPN account protected only by a password, with no multi-factor authentication behind it. One missing layer at the perimeter was enough to start a chain that shut down fuel distribution across the US East Coast for several days.

The Layers in Practice

Layers are usually described from the outside in, with two concerns cutting across all of them:

Notice that the list mixes preventive, detective, and corrective controls. That variety matters more than the number of layers.

Diversity Beats Repetition

Placing two identical firewalls back to back adds a layer on paper, but a single vulnerability or a single misconfiguration template can defeat both. Real depth comes from controls that fail for different reasons. A phishing email that gets past the mail gateway should still meet MFA on the account, least privilege on the endpoint, segmentation on the network, and an analyst watching for unusual behaviour.

Hidden Shared Dependencies

Layers that look independent are often tied together underneath. If your firewalls, EDR console, and backup server all authenticate against the same Active Directory and are managed from the same admin workstation, an attacker who captures one domain admin account can disable the lot. Ask of every layer: “If this falls, what else falls with it?” In most organisations the honest answer points to three places—the identity platform, the management plane, and the backup system. Protecting those separately, for example with dedicated admin accounts and backup credentials that sit outside the production domain, is what turns theoretical depth into real depth.

Defence in Depth and Zero Trust

The two ideas are complementary rather than competing. Classic layered designs had a weakness: once an attacker crossed the perimeter, the internal network trusted them far too much. Zero trust architecture removes that implicit trust by re-checking identity, device health, and context at every access decision. Modern designs keep the layers and apply zero trust principles at each boundary between them.

Costs, Limits, and How to Measure It

Adding controls does not automatically add security. Overlapping tools that nobody tunes generate false positives, slow teams down, and train analysts to ignore alerts. Every layer also carries licence, operating, and training costs. A sensible test is to map each control to the attacker techniques it stops or detects—MITRE ATT&CK is useful for this—and remove or consolidate those that add little.

The NCSC’s 10 Steps to Cyber Security is effectively a defence-in-depth programme for UK organisations, covering everything from network security and identity to logging, monitoring, and incident management. Cyber Essentials gives you a baseline for the most common layers, while assessments against frameworks such as ISO 27001 test whether those layers are governed consistently.

The most reliable way to find out whether your layers hold is to test them as an attacker would, end to end. At Privia Security, we help organisations measure that gap through our cyber risk analysis services.

FAQ – Frequently Asked Questions

How Many Layers Does Defence in Depth Need?

There is no magic number. What matters is that the controls protecting a critical asset fail for different reasons and that at least one of them can detect an attacker if the others are bypassed. Three well-run layers that are monitored and tested usually beat seven overlapping products that nobody tunes. Start from your most valuable data and systems and work outwards.

Is Defence in Depth Still Relevant in the Cloud?

Yes, although the layers look different. In AWS, Azure, or Google Cloud you rely on identity policies, network security groups, private subnets, encryption keys, and cloud-native logging rather than physical walls and appliances. The principle is unchanged: a single misconfigured storage bucket or leaked access key should not be enough to expose sensitive data on its own.

What Is the Difference Between Defence in Depth and Zero Trust?

Defence in depth is about stacking diverse controls so one failure does not become a breach. Zero trust is about never granting access based on network location alone and verifying every request. They work best together: zero trust tightens the boundaries between layers, while defence in depth makes sure there are several of them to cross.

How Can I Test Whether My Layers Actually Work?

Penetration tests check individual controls, while Red Team exercises and assumed-breach scenarios show whether an attacker can chain gaps across layers without being noticed. Purple Team sessions then let your defenders tune detections against the techniques that got through. Reviewing results against MITRE ATT&CK helps you see which layers are pulling their weight.

Related Terms

You May Be Interested In These

What Is Firmware Security?

Cybersecurity Terms

17 November 2026

What Is Firmware Security?

Read More
What Is a Malware Signature?

Cybersecurity Terms

17 November 2026

What Is a Malware Signature?

Read More
What Is a Bug Bounty Programme?

Cybersecurity Terms

17 November 2026

What Is a Bug Bounty Programme?

Read More