Guide Cybersecurity

What is cybersecurity?

Cybersecurity is the practice of protecting systems, networks, identities, applications and data from unauthorised access, attack, disruption and other digital threats. It involves technology, people, processes and management decisions. An effective programme protects the confidentiality, integrity and availability of information and prepares the organisation to detect, respond to and recover from incidents.

25 June 2026
Updated 10 November 2026
10 min read
Key takeaways
  • Cybersecurity protects systems, networks, identities, applications and data from unauthorised access, attack and disruption, and it covers people, processes and management decisions as well as technology
  • NIST CSF 2.0 organises a programme into six functions: Govern, Identify, Protect, Detect, Respond and Recover
  • In Verizon's 2026 DBIR, exploited vulnerabilities were the most common identified way into breached organisations (31%), ahead of credential abuse and phishing
  • Backups help you recover from ransomware, but recovery also needs containment, removal of attacker access and a check for stolen data
  • UK GDPR, the NIS Regulations 2018 and, for organisations operating in the EU, NIS2 set security and incident-reporting duties; whether each applies depends on your organisation

What cybersecurity covers

One way to explain the main technical controls is to group them into seven areas. The grouping is illustrative and the areas overlap. A complete programme also covers governance, risk management, suppliers, monitoring, incident response and recovery, which the next section describes.

1
Network security
Protecting the infrastructure that carries data between systems and to the internet: firewalls, intrusion detection, VPNs and network segmentation. These controls decide what traffic enters and leaves your environment and how far an attacker can move once inside.
2
Endpoint security
Protecting laptops, servers, phones and workstations. Endpoint detection and response (EDR) tools watch for malicious activity on each device, and each unmanaged device is a possible way in.
3
Application security
Securing the software you build and buy through secure development, code review, dependency scanning, penetration testing and web application firewalls. Publicly known, unpatched application flaws are a frequent entry point.
4
Cloud security
Securing workloads, data and identities in cloud services. Misconfigured storage, over-permissioned service accounts and thin logging are common sources of exposure. The provider secures its platform; your configuration choices are yours to get right.
5
Identity and access management
Controlling who and what can reach which systems. That includes staff accounts and machine identities such as API keys, service accounts and certificates. Multi-factor authentication, single sign-on, privileged access management and role-based access are the core tools.
6
Data security
Protecting sensitive data at rest and in transit with encryption, data loss prevention and access controls. It is one of the areas most directly tied to the security duties in UK GDPR.
7
AI security
Protecting AI tools and the data they use from attacks such as prompt injection, data poisoning and model theft, and using AI in detection and response. In IBM's 2026 study, malicious breaches classed as AI-enabled rose 56% on the previous period. That figure describes IBM's sample, not the total number of attacks worldwide.

The six functions of a security programme

The seven areas describe where controls sit. The NIST Cybersecurity Framework 2.0 describes how an organisation manages cybersecurity risk over time. Each of its six functions answers a question the board or leadership team should be able to answer.

1
Govern
Who is accountable for cybersecurity, and which risks are we prepared to accept?
2
Identify
Which systems, suppliers and information do we depend on, and which need protecting most?
3
Protect
Which safeguards reduce those risks to a level we accept?
4
Detect
How will we notice suspicious activity, and who looks at the alerts?
5
Respond
What happens when an incident occurs, who decides, and who must be notified?
6
Recover
How quickly can we restore critical services, and have we tested it?

Confidentiality, integrity and availability

Confidentiality, integrity and availability are three core properties of information security, often called the CIA triad. They help you judge what a control is for and what an incident has damaged. A full programme also considers authenticity, accountability, traceability and resilience.

C
Confidentiality
Information is available only to those authorised to see it. Encryption, access controls and least privilege protect it. A confidentiality breach means data reached someone who should not have it.
I
Integrity
Data is accurate and has not been altered without authorisation. Access controls and change management help prevent tampering; hashes, digital signatures and audit logs help detect it. A hash on its own does not stop an attacker who can change both the data and an unprotected hash.
A
Availability
Systems and data are there when the business needs them. Redundancy, tested backups, capacity planning and DDoS protection support availability. An availability breach means a service or data is out of reach when you need it.

The most common threats

$4.99M
average cost per breach across the organisations IBM studied (IBM 2026)
31%
of breaches with an identified entry point began with an exploited vulnerability (Verizon DBIR 2026)
48%
of breaches in Verizon's dataset involved ransomware (Verizon DBIR 2026)
1
Exploited vulnerabilities
Attackers use known flaws in internet-facing software such as VPNs, firewalls, file-transfer tools and web applications. Verizon's 2026 DBIR found this was the most common identified entry point, at 31% of the breaches it analysed. Fast patching of actively exploited flaws matters more than patching everything at the same pace.
2
Phishing and social engineering
Phishing remains a major technique for stealing credentials, committing payment fraud and installing malware. Business email compromise, spear-phishing and SMS phishing are variations. Verification steps for payments and changes to bank details reduce the damage when someone is fooled.
3
Ransomware and extortion
Criminals encrypt systems, steal data, or both, and demand payment. Some groups now skip encryption and only threaten to publish stolen data. Even when you recover without paying, the disruption, investigation and possible data exposure carry real cost. Fewer payments in some surveys do not mean fewer attacks.
4
Credential attacks
Credential stuffing (automated logins with passwords leaked elsewhere) and password spraying (a few common passwords tried against many accounts) target login portals. MFA blocks most automated attempts, though session theft and adversary-in-the-middle phishing can bypass weaker forms.
5
Supply chain attacks
Attackers compromise a supplier, software package or IT provider to reach its customers. Supplier risk reviews, contract security terms and a software bill of materials (SBOM) reduce exposure.
6
Insider threats
Staff, contractors and former employees with legitimate access can cause harm by mistake or on purpose. Least privilege, prompt removal of leavers' access and monitoring of privileged activity are the main controls.
7
AI-assisted attacks
Attackers use generative AI to write convincing phishing, clone voices and produce malicious code faster. They also target AI tools directly: prompt injection can make an assistant disclose data or take unintended actions.

The business and regulatory case

IBM's 2026 Cost of a Data Breach Report estimated an average cost of $4.99 million per breach across the organisations it studied. It also found average costs about $1.93 million lower where security AI and automation were used extensively, compared with organisations that used none. Both figures describe IBM's sample. They are not a forecast for a typical small business, and buying a tool does not guarantee that saving.

Patching, MFA and tested backups reduce important risks, but each protects a different stage of an incident. They work best inside a programme that also covers asset inventory, secure configuration, monitoring, incident response and supplier management. If you cannot hire a full security team, outside expertise such as a fractional security lead can set the priorities.

In the UK, UK GDPR requires appropriate security for personal data. When a personal data breach occurs, the controller must notify the ICO without undue delay and, where feasible, within 72 hours of becoming aware of it, unless the breach is unlikely to result in a risk to people's rights and freedoms. The maximum fine is £17.5 million or 4% of worldwide annual turnover, whichever is higher. The NIS Regulations 2018 add security and incident-reporting duties for operators of essential services and relevant digital service providers. See our guide to incident notification duties for the deadlines side by side.

UK reform and EU NIS2

The Cyber Security and Resilience Bill, introduced to Parliament in November 2025, would extend the NIS Regulations to more organisations, including managed service providers and data centres, and tighten incident reporting. If you operate in the EU, NIS2 may apply directly, depending on your sector, size and establishment under the relevant national law. Suppliers to NIS2 entities are not automatically in scope, but they often receive contractual security, continuity and notification requirements from their customers.

Common cybersecurity myths

1
"Strong passwords are enough"
A long password is harder to guess, but passwords are often stolen rather than guessed: exposed in other companies' breaches, captured by phishing or recorded by malware. MFA covers what a strong password cannot.
2
"We're too small to be a target"
Automated tools scan the internet for unpatched systems, exposed credentials and misconfigured services regardless of who owns them. Smaller organisations are found the same way as large ones, and they often have fewer people watching for an intrusion.
3
"Our industry isn't at risk"
Ransomware groups have hit councils, hospitals, schools, logistics firms and manufacturers as well as banks. Any organisation that relies on software and suppliers has supply chain exposure.
4
"We have antivirus, so we're protected"
Antivirus mainly detects known malware. Zero-day exploits, fileless techniques, stolen credentials and phishing pages that look legitimate can get past it. Antivirus is one layer of protection, not a security programme.
5
"We did an assessment, so we're covered"
Thousands of new vulnerabilities are published every year, and your own systems, staff and suppliers change too. An assessment is a snapshot. Reassess after significant changes and keep prioritising actively exploited vulnerabilities in between.

Where organisations start

Before choosing tools, identify your critical assets and cloud services, give named people responsibility for security, and agree who decides what happens during an incident. An incident response plan can be short to start with. On that base, these six controls close common gaps at reasonable cost:

1
One stolen password should not be enough to get in. Start with email, remote access, admin accounts and cloud consoles, and prefer phishing-resistant methods for privileged users.
2
Vulnerability management
Keep an inventory of what you run, set deadlines by severity, and patch internet-facing systems with actively exploited flaws first.
3
Backups with tested restores
Isolated or immutable backups, restored on a schedule, reduce your dependence on criminals to get systems back. Recovery also needs containment, removal of attacker access, credential resets and a check for stolen data. Our ransomware recovery guide covers the sequence.
4
Network segmentation
An attacker who reaches one system should not reach all of them. Segmentation limits lateral movement and keeps an incident contained.
5
Awareness and verification
Verizon's data shows people involved in a large share of breaches, through errors, phishing or misused credentials. Short, realistic training plus simple verification steps for payments and access requests work better than an annual tick-box course.
6
Monitoring your external exposure
Continuously discover internet-facing assets, such as old servers, forgotten subdomains and misconfigured cloud storage. External scanning complements an internal inventory; it does not replace it.

To sequence these alongside governance, budget and business priorities, a cybersecurity roadmap sets the order and the dependencies.

How Cyvra helps

Cyvra provides risk assessment, specialist cybersecurity consulting, security leadership, compliance preparation and support for implementing controls. Where you need continuous operations such as 24/7 monitoring, we help you select and coordinate the providers that run them. A cybersecurity assessment is the usual starting point. For sector-specific risks, see our guide to cybersecurity in hospitality.

Frequently asked questions

What is the difference between cybersecurity and information security?

Cybersecurity focuses on protecting digital systems, networks and data. Information security is broader and covers information in any form, including paper records and conversations. The terms are often used interchangeably, and most information security programmes today concentrate on digital risk.

What are the most common ways attackers gain initial access?

In Verizon's 2026 DBIR, exploitation of software vulnerabilities was the most common identified entry point, at 31% of the breaches analysed, followed by credential abuse. Phishing remains important, especially for stealing credentials and for payment fraud. Exposed remote access services with weak authentication are another frequent route.

Which UK laws apply to cybersecurity?

UK GDPR requires appropriate security for personal data and, for breaches likely to pose a risk to individuals, notification to the ICO within 72 hours of becoming aware where feasible. The NIS Regulations 2018 apply to operators of essential services and relevant digital service providers, and the Cyber Security and Resilience Bill would extend them. Financial services firms also face FCA and PRA operational resilience rules. If you operate in the EU, NIS2, GDPR and DORA may apply as well.

Is ISO 27001 a legal requirement?

Not for every organisation. ISO/IEC 27001 is a voluntary standard, but some contracts, tenders and sector schemes require certification or equivalent security evidence. Certification supports a security management programme; it is not proof of UK GDPR compliance and does not replace a legal assessment.

How often should a security assessment be conducted?

It depends on your exposure, how critical your systems are, how often they change, and any legal or contractual obligations. PCI DSS, for example, sets its own scanning and testing intervals. Reassess risk after significant changes, and prioritise actively exploited vulnerabilities continuously rather than waiting for an annual review.

Is cybersecurity only relevant for large organisations?

No. Automated tools scan the whole internet for vulnerable systems, whatever the size of the owner. Smaller organisations often have fewer controls and fewer people watching for an intrusion, so an attack can go unnoticed for longer. The NCSC's Cyber Essentials scheme sets a baseline that suits many small businesses.

How does AI affect cybersecurity?

Attackers use generative AI to write convincing phishing, clone voices and produce malicious code faster. Defenders use it for detection and triage. Your own AI tools need protecting too, because prompt injection and data poisoning target the AI systems themselves.

Sources

Ryland Deakin
About the author
Lead Consultant, Cyvra · CISM · CompTIA Security+ · MCP

Ryland has delivered cybersecurity, compliance, and IT management programmes for regulated organisations across the UK and the Netherlands for over 20 years. View full profile

Talk to Cyvra

Assess your organisation's real risks

We identify your critical systems and services, priority vulnerabilities, access risks and supplier dependencies. You get an assessment with recommended actions, priorities and owners, sized to your organisation and its obligations.

Disclaimer: This article is for general informational purposes only and does not constitute legal, regulatory, or professional advice. Cyvra makes no warranty as to the accuracy or completeness of this content. Readers should seek independent advice appropriate to their specific circumstances. Cyvra accepts no liability for any loss arising from reliance on this content.