Human risk management
Most breaches involve a person. Train little and often, and make reporting easy.
A once-a-year training video does not change behaviour. Short regular training, fair phishing practice, fast reporting and controls that catch mistakes do. Here is what a human risk programme looks like, how to run it without blaming people, and a 9-point check for your business.
Breach data from the Verizon 2026 DBIR, checked 31 October 2026.
- What a programme includes
- Culture, not blame
- 9-point check
Reviewed by Blake Bath, General Manager, on 31 October 2026.

The risks
Four ways people end up in a breach.
None of them need a bad employee. Most need a busy one.
Phishing
A convincing email, often AI-written, gets a password or a payment.
- Simulations
- Report button
- Phishing-resistant MFA
Human error
The wrong attachment or recipient exposes personal information.
- Short training
- Email warnings
- Clear rules
Reused passwords
A password leaked from another site opens a work account.
- Password manager
- Dark web monitoring
- MFA
No clear rules
Nobody knows who to tell, or what is allowed on a personal phone.
- Short policies
- Signed
- Known contact
Human risk check
How well does your business manage human risk?
Tick what is true today. You get the gaps, in the order we would close them.
Tick what is true for your business today.
Where you stand
0of 9 practices in place
9 gaps to close.
- Short training every month, not once a yearPeople forget annual training. Little and often sticks.
- Regular phishing simulationsPractice is how people learn to spot the real thing.
- A one-click way to report suspicious emailFast reports turn one click into a contained incident.
- Reporting a mistake is thanked, not punishedBlame teaches people to hide mistakes.
- Leaked staff passwords are watched forReused passwords from other breaches are a common way in.
- MFA backs up the peopleTraining lowers the odds. Controls limit the damage.
- Payment changes always get a callEmail fraud targets people, not systems.
- Policies are short, signed and knownClients, insurers and SMB1001 ask for the evidence.
- Risk is measured over timeIf you do not measure it, you cannot show it is improving.
What it includes
What a human risk programme includes.
- Short, regular trainingA few minutes a month, matched to each person's role and results.
- Phishing simulationsRealistic and varied, with instant private follow-up for anyone who clicks.
- Leaked password monitoringWatching for staff credentials in breach data and resetting them.
- Policies people signShort policies with signatures tracked.
- A human risk scoreOne measure per person and team, trending over time.
Not blame
Build a culture where people report, not hide.
- Thank reports, even late onesSomeone who clicked and reported has done the most useful thing.
- Leaders go firstExecutives do the same training and simulations.
- Keep results privateIndividual results go to the individual, team trends go to managers.
- Explain the whyReal local examples, like a changed invoice, land better than generic warnings.
Evidence
Where training fits with frameworks and insurers.
The Essential Eight does not include training as one of its eight strategies, but training supports controls such as MFA and macro settings by reducing how often they are tested. SMB1001 includes staff training requirements, and many insurers and client questionnaires ask for training records, simulation results and signed policies.
A managed programme produces that evidence as it runs, so you are not scrambling for completion records when an insurer or client asks.
How we run it
How we run it.
We run our managed human risk service on usecure. It covers training, phishing simulations, policy management and dark web monitoring, syncs users from Microsoft 365 or Google Workspace, and reports on each person's risk over time.
We set it up, tune the simulations to your business, review the results with you each quarter, and act on what they show.
Our view
Training lowers the odds. Controls limit the damage.
No training programme gets clicks to zero, and it should not have to. Pair it with phishing-resistant MFA, payment call-backs and someone watching sign-ins, so the one mistake that does happen is caught before it costs money.
Then measure. The businesses that improve fastest are the ones that look at the numbers every quarter and adjust.
The order
Starting a human risk programme in five steps.
- Step 1Baseline
A first simulation and leaked password check.
- Step 2Set up training
Short, monthly, matched to roles.
- Step 3Make reporting easy
A report button and a no-blame message from leadership.
- Step 4Sign the policies
Short policies, signed and tracked.
- Step 5Review quarterly
Look at the trend and adjust.
Related guides
People are one layer. Here are the others.
Phishing, email fraud, sign-in and policies.
Get help your way
Six ways to reach us. Pick whichever suits you.
Helpdesk hours are Monday to Friday, 8am to 5pm AEST. On Advantage and above, emergencies get a response within 1 hour and outages within 2, with 24/7 best-effort escalation.
- Call 07 3114 2808
An engineer answers Monday to Friday, 8am to 5pm AEST. After hours, weekends and public holidays, Chip, our AI agent, answers on the same number or on 07 3180 0965, logs it and escalates urgent incidents to the on-call engineer.
Call now - Email the helpdesk
Send it to helpdesk@realbytes.com.au and a ticket is logged straight away, with a reference number you can quote.
Email us - Client portal
Log and follow your tickets, see what we are working on and who has it.
Open the portal - Remote support
When an engineer asks to see your screen, start a secure session here. You start it, so nobody sees your screen until you do.
Start a session - Chat in Microsoft Teams
Clients can message Chip, our AI assistant, in Teams to log a request without leaving their desk.
How clients reach us - Service status
Check whether Microsoft 365 or another cloud service is down before you lose an hour troubleshooting.
See live status
Questions
What people ask about human risk management.
What is human risk management?
An ongoing programme that measures and reduces the risk that comes from people: short regular training, phishing simulations, watching for leaked passwords, signed policies, and a culture where people report mistakes quickly. It replaces the once-a-year training session.
How is it different from security awareness training?
Awareness training is one part of it. Human risk management adds measurement, so you can see who is at risk and whether it is improving, and ties training to what each person actually does and gets wrong.
How much of the risk comes from people?
In the Verizon 2026 Data Breach Investigations Report, 62% of breaches involved the human element, such as a click, a mistake or a misused account.
Does the Essential Eight require staff training?
No. Training is not one of the Essential Eight strategies, although it supports them. Other frameworks and many insurers do ask for evidence of training, simulations and signed policies, and SMB1001 includes staff training requirements.
Will phishing simulations upset staff?
Not if they are run fairly. Keep results private, give instant short training rather than public shaming, and celebrate reports. The aim is practice, not catching people out.
What platform do you use?
We run our managed human risk service on usecure, which covers training, phishing simulations, policy management and dark web monitoring, and syncs users from Microsoft 365 or Google Workspace.
Next step
Want a programme that changes behaviour?
We set up training, simulations, leaked password monitoring and policies, run them for you, and show you the trend every quarter.
Keep reading
Reviewed by Blake Bath, General Manager, on 31 October 2026. Breach data from the Verizon 2026 DBIR. Framework details from ASD. Checked 31 October 2026. Positive Technologies, Verizon DBIR 2026 summary; ASD, Essential Eight maturity model; ASD, Learn about phishing scams; OAIC, Notifiable Data Breaches scheme

Remote