17 November 2026 • Cyber security
In brief: Human risk management (HRM) is an approach to cybersecurity that measures how people actually behave, works out where the human risk sits, and then uses targeted training, simulations and process changes to bring that risk down. The term was popularised by the analyst firm Forrester, which presented it in 2022 as the successor to classic security awareness training. The difference is the starting point: awareness training asks whether everyone completed the course, human risk management asks whether behaviour changed and risk went down. You start with a baseline, a few behaviours that matter, short and frequent practice, and reporting that management can act on.
What is human risk management?
Forrester’s analyst Jinan Budge introduced the idea in 2022 and formally defined the market in 2024. In that definition, human risk management covers “solutions that manage and reduce cybersecurity risks posed by and to humans” through four things:
- detecting and measuring human security behaviours and quantifying the human risk;
- initiating policy and training interventions based on that risk;
- educating and enabling the workforce to protect themselves and the organisation;
- building a positive security culture.
Training is still in there. It just stops being the goal and becomes one of the tools. In its 2026 landscape, Forrester widened the definition to include behaviour around AI systems and agents, and noted that buyers are moving past completion rates and simulation scores towards the question of whether risk really fell.
Gartner describes a related idea under a different name: the security behaviour and culture program (SBCP). The labels differ, the shift is the same. From telling people the rules to changing what they do on a normal working day.
Why people are the risk worth managing
Most attacks still go through a person. According to the Verizon Data Breach Investigations Report, the human element was involved in 62% of breaches in the 2026 edition, through phishing, stolen credentials, social engineering and plain mistakes. In Europe, ENISA’s Threat Landscape 2025 found that phishing, including vishing and malvertising, accounted for about 60% of the observed initial intrusions.
Those numbers are not new, and that is the point. Many organisations have run awareness training for years, and the share of incidents that start with a person barely moves. A module once a year and a completion certificate do not change how someone reacts to a convincing invoice email on a busy Friday afternoon.
Security awareness training versus human risk management
The two are not opposites. HRM builds on awareness training, but it changes what you measure and how often you act.
| Classic security awareness training | Human risk management | |
|---|---|---|
| Focus | Knowledge: do people know the rules? | Behaviour and risk: do people act safely, and where is the risk highest? |
| Measurement | Completion rates, quiz scores | Behaviour over time: simulation results, reporting rate, repeat mistakes, trends per team |
| Frequency | Annual or onboarding module | Continuous: short, frequent moments of practice |
| Personalisation | Same course for everyone | Content and follow-up based on role, department and observed behaviour |
| Data | Data from the learning platform only | Training and simulation data, plus incident reports and signals from security tools where available |
| Outcome for management | Proof that training took place | Insight into where risk sits and whether it is going down |

The building blocks of a human risk approach
Measuring risk per person and per team
You cannot manage what you have not measured. A human risk approach starts with a picture of the current situation: who clicks, who reports, which departments struggle with which topics. A finance team that receives payment requests every day carries a different risk than a warehouse team without a mailbox. Measuring per department keeps the effort where it matters.
Influencing behaviour with short, frequent practice
Behaviour changes through repetition, not through one long session. Short training moments spread over the year keep the topic in mind without taking people away from their work. Phishing simulations add realistic practice: employees learn to recognise a fake message in a safe setting, and you see whether the training sticks. No law requires simulations. They are simply one of the few ways to observe behaviour rather than knowledge.
A culture in which people report
The most valuable behaviour is not “never click”. It is “report quickly, also when you did click”. That only happens when employees are not punished or named for a mistake. A good security awareness culture makes reporting normal and rewards it, so the security team hears about an attack within minutes instead of days.
Reporting that leads to decisions
Data only helps when someone acts on it. Management wants to know whether the risk is going down, the security team wants to know which department needs a follow-up session, and an auditor wants evidence. Reporting should answer those questions in plain language, without turning into a list of names.
Which data a human risk approach uses
Mature HRM tools combine many signals. You do not need all of them to start. Roughly, there are three layers.
- Training data: participation, progress, scores per topic and per department.
- Behavioural data: results of phishing simulations, such as who opens, clicks, enters data or reports the message, and how that changes over time.
- Operational signals: real phishing reports to the service desk, incidents caused by human error, MFA enrolment, risky sign-ins or data loss alerts from existing security tools.
Context makes the data meaningful. An employee with administrator rights or access to payments weighs more heavily than someone without. Start with the first two layers, which most organisations already have or can set up quickly, and add operational signals once the basics run.
Human risk management starts with training that people actually do. See how Guardey's 3-minute weekly challenges and phishing simulations show progress per employee and department.
How to start with human risk management
- Measure a baseline. Run a first phishing simulation and a short knowledge check before you change anything. Our guide to a security awareness baseline measurement explains how.
- Choose three to five behaviours that matter. For example: reporting suspicious emails, checking payment changes by phone, not sharing MFA codes, locking screens. Fewer goals make progress visible.
- Agree on privacy and governance first. Decide which data you collect, who sees results at individual level and how long you keep them. Involve HR, the privacy officer and, where there is one, the works council.
- Replace the annual module with a rhythm. Short training every week or every few weeks, plus regular simulations that vary in difficulty and channel.
- Make reporting easy and visible. One button or one address to report suspicious messages, and a quick thank you from the security team.
- Follow up per team, not per culprit. Use the results to give a department extra training on a weak topic, or to change a process that invites mistakes.
- Report to management every quarter. Show the trend in reporting rate, click rate and participation, and the actions you took.
Pitfalls to avoid
Treating behavioural data per employee lightly
Human risk management is about people, and its data is personal data. Simulation results and risk scores per employee fall under the GDPR. Collect only what you need for the security goal, be transparent towards employees and limit who sees individual results.
Turning measurement into blame
A public list of who clicked the most is the fastest way to kill your reporting culture. People who fear being named stop reporting their mistakes, and then you lose the signal that matters most.
Staring at the click rate
A click rate depends heavily on how difficult the simulation was. A falling click rate after easy tests says little. The reporting rate, and how fast reports come in, tells you more about real resilience.
Buying a tool before you have goals
A platform with dashboards does not create a programme. Decide first which behaviours you want to change and who acts on the results. Then choose the tooling.
Human risk management and NIS2
The NIS2 Directive does not mention human risk management. It does contain two obligations that a human risk approach helps to meet. Article 21(2)(g) lists basic cyber hygiene practices and cybersecurity training among the minimum measures, and Article 21(1) asks for measures that are proportionate to the risk. Article 20(2) requires members of the management body to follow training and encourages entities to offer similar training to employees regularly. HRM is not a legal requirement. It is a way to show that training is risk-based and that it works, which is the kind of evidence a supervisor or auditor tends to ask for.
Where Guardey fits
Guardey offers security awareness training in three-minute weekly challenges, with gamification such as points and a leaderboard to keep participation high. Phishing simulations let employees practise spotting fake messages in a safe environment. Real-time analytics and weekly reports show participation and progress per employee and department, which topics score lowest, and which colleagues may need extra training. That covers the measurement, practice and reporting blocks of a human risk approach.
Want to compare tools that position themselves as human risk platforms? Our overview of human risk management platforms lists the options side by side.
Frequently asked questions about human risk management
What is human risk management in cybersecurity?
Human risk management is an approach that measures security behaviour of employees, determines where the human risk is highest and reduces it with targeted training, simulations, process changes and a positive security culture. The term was popularised by Forrester as the successor to classic security awareness training.
What is the difference between security awareness training and human risk management?
Security awareness training focuses on knowledge and usually measures completion. Human risk management focuses on behaviour and risk: it measures how people act over time, tailors follow-up to roles and teams, and reports whether risk goes down.
Is human risk management required under NIS2?
No. NIS2 requires cybersecurity training, cyber hygiene and training for management, and measures that fit the risk. It does not prescribe human risk management. A human risk approach is one way to show that your training is risk-based and effective.
Can you measure human risk per employee under the GDPR?
Yes, but with care. Results per employee are personal data. Collect only what you need, inform employees, restrict access to individual results and, in countries with works councils such as the Netherlands, involve the works council before you start.
In a personal demo we show how the reports per employee and department point to where extra training is needed.