Security awareness training that changes behaviour

Line drawing of a laptop with an envelope lifting off the screen towards a raised flag, the flag in blue

Most programmes produce completion certificates, not safer habits. This guide shows how to design training around the few behaviours that actually reduce risk, and how to tell whether it works.

In short

Security awareness training changes behaviour when it targets a handful of specific actions, such as reporting suspicious messages, using MFA and handling data correctly, delivers short lessons inside daily work, adapts content to each role, and measures reporting rates and response times rather than course completion. Pair it with technical controls, because training cannot replace them.

Annual slide decks and a quiz rarely change what people do on a busy Tuesday. This guide is written by Devies, a Gothenburg-based product and engineering consultancy, for IT leaders, CISOs and HR or compliance owners who are responsible for security awareness training (in Swedish often called säkerhetsutbildning) and want it to reduce incidents rather than tick a box. It covers what effective training targets, a step-by-step method, how to run phishing exercises without damaging trust, and how to measure results.

What should security awareness training actually change?

Security awareness training should change a small number of observable behaviours, because knowledge alone does not reduce incidents and people cannot act on fifty rules at once.

A few definitions keep the scope honest:

  • Awareness is knowing that a risk exists, for example that invoices can be faked.
  • Training is practising a specific response, for example calling the supplier on a known number before changing bank details.
  • Security culture is the shared expectation that reporting a mistake is welcome and quick.

Pick five to eight target behaviours that match your real incidents. Typical candidates:

  1. Report suspicious emails, calls and messages through one simple button or channel.
  2. Verify payment and bank detail changes through a second channel.
  3. Approve MFA prompts only when you started the sign-in yourself.
  4. Use the password manager and never reuse work passwords.
  5. Lock screens and devices, and report lost equipment the same day.
  6. Share files through approved tools, not personal accounts.
  7. For developers: never commit secrets, and follow the secure code review checklist.

NIST SP 800-50 Revision 1, published in September 2024, frames this as a cybersecurity and privacy learning programme with a life cycle: plan, design, develop, implement and assess. Its key idea is that learning is ongoing and role-based, not a yearly event.

How do you build a programme step by step?

Build the programme from your own incident data outwards: first decide which behaviours matter, then who needs them, then how and how often to train.

Circular path of eight stepping stones from incident evidence to measurement, looping back to the start
  1. Collect evidence. Review the last year of incidents, near misses, help desk tickets and phishing reports. Ask which human actions would have stopped or shortened them.
  2. Choose target behaviours. Write each as an action with a trigger: "When an email asks you to change payment details, call the supplier on the number in the ERP system."
  3. Segment by role. Finance, HR, developers, administrators, customer support and executives face different attacks. Give each group a core module plus role-specific content.
  4. Remove friction first. If reporting takes six clicks, fix the tool before training people to use it. Add a report button to the mail client and confirm every report with a thank-you.
  5. Deliver little and often. Five-minute lessons each month beat a two-hour annual course. Trigger short lessons at relevant moments: onboarding, first payment approval, a new admin role.
  6. Practise with realistic exercises. Use simulations and tabletop exercises that mirror real scenarios, including phone and messaging attacks, not only email.
  7. Train leadership separately. Management needs to understand risk decisions, reporting duties and their own exposure to targeted fraud.
  8. Measure and adjust quarterly. Track the metrics below, retire content that does not move them, and add topics as new attack patterns appear.

How should you run phishing exercises without losing trust?

Phishing exercises work when they teach reporting, not when they catch people out, so the main goal is a high and fast report rate, not a low click rate.

Phishing training (phishing utbildning) is the part most organisations start with, and the part most often done badly. The UK National Cyber Security Centre's phishing guidance warns that relying on users to spot every phishing email is unrealistic and recommends a multi-layered defence: filtering, protected devices and accounts, plus a culture where people report without fear. Its advice also cautions against using simulations to blame individuals.

Practical rules:

  • Announce the programme in general terms. People should know exercises happen, not when.
  • Never shame clickers. Show a short, friendly explanation at the moment of the click, and never publish names.
  • Celebrate reports. A quick report from one person can protect everyone else.
  • Vary difficulty and channel. Include SMS, voice and collaboration tools, since attackers do.
  • Avoid cruel lures. Fake bonuses or health scares damage trust in internal communication for months.

How do you measure whether the training works?

Measure behaviour and response speed, because completion rates show only that people clicked through a course, not that they act differently.

Metric

What it tells you

Direction you want

Report rate in simulations

What it tells you

Whether people use the reporting channel

Direction you want

Up

Median time to first report

What it tells you

How quickly the security team learns of an attack

Direction you want

Down

Real suspicious-message reports per month

What it tells you

Whether the habit transfers to real attacks

Direction you want

Up, then stable

Repeat clicks by the same person

What it tells you

Whether support reaches those who need it

Direction you want

Down

Incidents with a human trigger

What it tells you

Whether the training touches real risk

Direction you want

Down

Completion rate

What it tells you

Only whether the course was delivered

Direction you want

Necessary, not sufficient

Read the numbers together. A falling click rate with no reports may mean the simulations became easier, not that people became safer. Fast reports feed directly into your incident response plan, so agree with the security team how a reported message is triaged and how the reporter gets feedback.

How does training fit with NIS2 and technical controls?

Training is required for some organisations and useful for all, but it works only alongside technical controls that limit the damage of the mistakes people will still make.

As of October 2026, Article 20(2) of NIS2, Directive (EU) 2022/2555 requires members of the management bodies of essential and important entities to follow training, and encourages entities to offer similar training to employees regularly. Article 21 also lists basic cyber hygiene practices and cybersecurity training among the risk-management measures. In Sweden these duties sit in the Cybersecurity Act (cybersäkerhetslagen); see what NIS2 and the Swedish Cybersecurity Act mean in practice.

Training also makes more sense when the safe path is the easy path. Phishing-resistant MFA, single sign-on and least privilege mean that a stolen password or an approved prompt causes far less harm; read identity and access management explained for the technical side. A useful rule: for every behaviour you train, ask which control would catch the failure, and fund both.

Devies

Want a second opinion?

Devies, a Gothenburg-based product and engineering consultancy, works on questions like this with Nordic product and IT teams. Tell us where you are and we will suggest a sensible first step.

Talk to Devies

Frequently asked questions

How often should employees receive security awareness training?

Short, regular sessions work better than a single annual course. A practical rhythm is a brief lesson or exercise every month, an onboarding module for new starters, role-specific modules for finance, administrators and developers, and an annual refresher for management. Add just-in-time lessons when someone gets a new role or privilege. Review the content each quarter against current incidents and attack patterns.

Is security awareness training mandatory under NIS2?

NIS2 makes training mandatory for members of management bodies at essential and important entities, and requires entities to encourage regular training for employees. It also lists cyber hygiene and cybersecurity training among the required risk-management measures. In Sweden the rules apply through the Cybersecurity Act. Organisations outside NIS2 scope have no general legal duty, but customers and insurers increasingly ask for evidence of training.

Do phishing simulations work?

Phishing simulations help when they are used to practise reporting and to give friendly feedback at the moment of a mistake. They do harm when used to shame or punish people, because staff then hide mistakes, which delays incident response. Measure report rate and time to report rather than click rate alone, vary the channels and difficulty, and combine simulations with technical controls such as filtering and phishing-resistant MFA.

What should developers learn in security training?

Developers need practical, code-level training rather than generic awareness: handling secrets, validating input, using authentication and authorisation libraries correctly, reviewing dependencies and reading security findings from the pipeline. Base the content on the OWASP Top 10 and on vulnerabilities found in your own code. Short exercises in the team's own language and frameworks, plus a cybersecurity champion in each team, tend to stick better than video courses.

Sources

Here’s more

Join Our Newsletter

We handle your data according to our privacy policy.

Shopping Cart

We use cookies

We use cookies to improve your experience on this website. You may choose which types of cookies to allow and change your preferences at any time. Disabling cookies may impact your experience on this website. You can learn more by viewing our Cookie Policy.

Check out

Contact Form text

We will contact you text

We handle your data according to our privacy policy.