Data Breach
About 4 min read
Last updated: 2026-08-24
What Is a Data Breach
A data breach is an event in which information that should be protected is accessed or obtained by a third party without authorization. The types of data exposed vary widely - personal information, credentials, financial data, and corporate trade secrets.
Data breaches are caused not only by external attacks but also by internal human error and misconfiguration. IBM and Ponemon Institute's Cost of a Data Breach Report 2026 puts the global average cost of a single data breach at 4.99 million US dollars (up 12 % year over year), citing the burden of detection and response and lost business opportunities as drivers of the increase. A breach delivers a triple blow to an organization: financial loss, damage to trust, and legal liability. Leaked credentials are also abused in credential stuffing, so the damage tends to spread in a chain.
Main Causes and Attack Vectors
The causes of data breaches fall into three broad categories.
External Attacks
- Phishing and Social Engineering: Employees are tricked into handing over credentials, which are then used to enter internal systems. Because the attacker logs in with valid credentials, traces of the intrusion tend to be buried in ordinary business access
- Exploitation of Vulnerabilities: Unpatched software or configuration gaps are used to get in. SQL injection in web applications and inadequate authentication on APIs are frequent targets
- Ransomware: Under the double extortion pattern, where the threat of publishing stolen data is added to encryption of the data, recovering from backups does not undo the exposure itself
Internal Factors
- Misconfiguration: Mistaken access permission settings on cloud storage, databases left open to the public, and similar gaps. No intrusion traces are left behind, so the organization cannot notice it in its own logs and sometimes learns of it only when an outside party points it out
- Insider Misconduct: Data carried out by employees about to leave, or unauthorized access through abuse of privileges
Through Third Parties
- Compromise of a contractor or business partner: Even if your own security is solid, a breach at a partner you share data with produces a leak. See supply chain attack
The longer the dwell time between intrusion and discovery, the harder it becomes to establish after the fact what was taken. Activity beyond the log retention period cannot be traced, and an organization can end up having to decide the scope of notification without being able to confirm whether data was exposed at all.
Incident Response Procedures
When a data breach occurs, swift action based on an incident response plan limits the spread of damage.
- Containment: Isolate the compromised systems from the network and cut off the attacker's access. If credentials were exposed, reset the passwords immediately
- Establishing the scope: Investigate which data was exposed, over what period, and to what extent. Analyze logs and carry out forensic examination
- Notification: Article 33 of the GDPR requires notification to the supervisory authority without undue delay and, where feasible, within 72 hours after becoming aware of the breach (with an exception where a risk to rights is unlikely). Under Japan's Act on the Protection of Personal Information (amended in 2020, in force from April 2022), reporting to the Personal Information Protection Commission and notifying the affected individuals become mandatory when sensitive personal information is involved, when financial damage may result, when the act was carried out for fraudulent purposes, or when more than 1,000 individuals are affected
- Recovery: Identify and close the intrusion path, then restore systems to a safe state. Recover from backups, apply patches, and review configurations
- Post-incident analysis and prevention of recurrence: Analyze the root cause and identify where security measures need to improve. Add monitoring rules that detect the same class of attack
Prevention and Damage Mitigation
Preventing data breaches completely is difficult, but lowering the probability of occurrence and minimizing the damage are achievable.
Preventive Measures
- Data encryption: Encrypt data both at rest and in transit. Even when a breach occurs, encrypted data is far less valuable to the attacker
- Organization-wide two-factor authentication: Even if a password leaks, an account cannot be taken over without the second authentication factor. It is essential for administrator accounts in particular
- Principle of least privilege: Grant users and systems only the minimum access rights they need, limiting how far a compromise can reach
- Regular vulnerability scanning and penetration testing: Find and fix weaknesses before an attacker does
Damage Mitigation
- Data classification and minimization: Hold only the data you need and apply protection levels according to sensitivity. Delete data reliably once it is no longer needed
- Network segmentation: Divide the network logically so that a breach of one segment does not spread laterally to the others
- Log aggregation and monitoring: Deploy SIEM (Security Information and Event Management) to detect abnormal data access patterns in real time
These measures divide into a layer that keeps intruders out and a layer that reduces what can be carried away once someone is inside. Stacking up only the former means the damage peaks the moment it is broken through, so it is important to also prepare measures such as encryption and segmentation that determine what remains after the barrier falls.
Common Misconceptions
- Data breaches are only a problem for large enterprises
- Small and medium businesses are actually easier targets for attackers due to limited security investment. They may also be used as stepping stones to larger enterprises through the supply chain. The risk exists regardless of company size.
- If your password is leaked, just changing it makes you safe
- If the leaked password was reused across services, all of them need to be changed. Additionally, leaked personal information (name, address, date of birth) can be used to crack security questions or improve phishing accuracy - password changes alone are insufficient.