🕑 Estimated reading time: 4 minutes | Last updated: June 11, 2024
Zero trust is a security model that assumes no user, device or network location is automatically trustworthy. Every access request is verified explicitly using identity, device health and context; access is granted with least privilege; and systems are designed assuming a breach may already have happened. Defined in NIST SP 800-207 and promoted by the US government and CISA’s Zero Trust Maturity Model, it is implemented step by step across identity, devices, networks, applications and data.

The old idea of security was a strong wall around the office network. But staff now work from home, apps live in the cloud and attackers routinely get past the wall. Zero trust starts from a different question: why trust anyone by default?
For decades, corporate security looked like a castle with a moat. Firewalls protected the network perimeter, and anyone inside, employees in the office or devices on the internal network, was largely trusted. That model has broken down. Staff work from home and on mobile devices, applications run in the cloud, partners connect from outside, and attackers who get past the perimeter through a single phishing email can often move freely inside.
Zero trust is the response. Its motto, “never trust, always verify,” reflects a simple idea: trust should never be granted automatically based on where a request comes from.
Table of Contents
The Core Principles
The US National Institute of Standards and Technology describes zero trust in its Special Publication 800-207, published in 2020. Most frameworks share three principles:
- Verify explicitly. Authenticate and authorise every request using all available signals: user identity, device health, location, time, the sensitivity of the resource and behaviour patterns.
- Use least privilege access. Grant only the access needed for a specific task, for a limited time where possible (“just-in-time” and “just-enough” access).
- Assume breach. Design systems as if an attacker is already inside: segment networks, encrypt traffic, monitor continuously and limit the “blast radius” of any compromise.
Why Zero Trust Has Become a Priority
High-profile incidents, such as supply chain attacks and ransomware outbreaks that spread across flat internal networks, showed the weakness of perimeter-based security. In May 2021, the US government’s Executive Order 14028 on improving the nation’s cybersecurity directed federal agencies to adopt zero trust architectures, followed by a federal zero trust strategy in January 2022. Many private organisations and governments elsewhere have followed.
The Five Pillars
The US Cybersecurity and Infrastructure Security Agency’s Zero Trust Maturity Model (version 2.0, published in 2023) organises zero trust into five pillars, each with maturity stages from traditional to optimal:
1. Identity
Strong identity is the foundation. This means centralised identity management, phishing-resistant multi-factor authentication, single sign-on, and continuous evaluation of risk during sessions, for example re-checking when a user suddenly logs in from a new country.
2. Devices
Organisations must know which devices are accessing resources and whether they are healthy: managed, encrypted, patched and running endpoint protection. Access can be limited or denied for unmanaged or compromised devices.
3. Networks
Instead of one large trusted network, zero trust uses micro-segmentation to isolate workloads, encrypts traffic internally as well as externally, and replaces broad VPN access with application-specific access.
4. Applications and workloads
Applications should authenticate and authorise every user and service, be tested for vulnerabilities, and be accessible only through secure gateways rather than exposed directly to the internet.
5. Data
Data is classified and protected according to sensitivity, with encryption, access controls and monitoring for unusual access or exfiltration.
Across all pillars run three capabilities: visibility and analytics, automation and orchestration, and governance.
Zero Trust Network Access vs VPN
Traditional VPNs typically place remote users onto the internal network, from where they can reach many systems. Zero trust network access (ZTNA) instead connects users only to the specific applications they are authorised to use, after verifying identity and device posture, and hides other resources entirely. If an account is compromised, the attacker’s reach is far more limited.
Common Misconceptions
- “Zero trust is a product.” It is a strategy and architecture. Products can help, but no single tool delivers zero trust.
- “Zero trust means we don’t trust our employees.” It means the system does not assume trust; it continuously verifies, which protects employees whose credentials may be stolen.
- “It is only for large enterprises.” Small organisations can apply the same principles using cloud identity services, MFA, device management and least privilege.
- “We can do it in one project.” Zero trust is a journey that usually takes years.
How to Get Started
- Identify your protect surface: the critical data, applications, assets and services that matter most.
- Map how they are accessed: which users, devices and systems need to reach them.
- Strengthen identity with single sign-on and phishing-resistant MFA.
- Get visibility of devices and enforce basic health requirements.
- Reduce standing privileges, especially administrator access.
- Segment critical systems and replace broad network access with application-level access.
- Monitor and improve continuously, using logs and analytics to refine policies.
The Benefits
- reduced impact of stolen credentials and phishing;
- better support for remote and hybrid work;
- improved visibility of who is accessing what;
- easier compliance with regulations that require strong access control;
- a smaller blast radius when incidents occur.
Verify access wherever people connect
Zero trust reflects how organisations actually work today: in the cloud, on mobile devices and across many networks. By verifying every request, limiting privileges and assuming breach, it turns security from a wall around the network into protection that follows users and data wherever they go.
Frequently Asked Questions
Is zero trust the same as MFA?
No. Strong MFA is an essential part of zero trust, but the model also covers devices, networks, applications and data.
Does zero trust eliminate the need for firewalls?
No. Firewalls still play a role, but they are no longer the main source of trust.
How long does zero trust take to implement?
Most organisations adopt it in phases over several years, starting with identity and the most critical assets.