Initializing secure session…
Now enrolling — Compliance-as-a-Service & vCISO-as-a-ServiceLimited to 30 client seats this seasonAgentic AI · Robotics · Zero Trust CybersecurityServing North America · Africa · AsiaFree Security Readiness Call — no obligationNIST CSF 2.0 · CMMC · HIPAA · GDPR · SOC 2 · ISO 27001Now enrolling — Compliance-as-a-Service & vCISO-as-a-ServiceLimited to 30 client seats this seasonAgentic AI · Robotics · Zero Trust CybersecurityServing North America · Africa · AsiaFree Security Readiness Call — no obligationNIST CSF 2.0 · CMMC · HIPAA · GDPR · SOC 2 · ISO 27001
Training · 12 Oct 2026 · 5 min read

Security training that changes behavior: why completion rates are the wrong metric

Ninety-eight percent of staff completed the annual module. The phishing click rate did not move. If that sounds familiar, the program is measuring the wrong thing.

Most security-awareness programs are designed to produce a number: the percentage of staff who completed the training. Auditors accept it, insurers ask for it and leadership is reassured by it. It tells you almost nothing about whether the organisation is safer.

What the number hides

Completion measures attendance, not behavior. A nurse who clicked through forty slides between patients, a driver who watched a video on a phone at a loading dock, a partner whose assistant completed it for them — all count as trained. None of them will necessarily pause before approving a wire transfer or opening an attachment from a "supplier."

The attacks that succeed against organisations are overwhelmingly human: credential phishing, business-email compromise, pretexting calls to the help desk. Training that does not change what people do in those moments is a compliance artefact, not a control.

A better set of questions

Did the phishing-simulation click rate fall? Did reporting of suspicious messages rise? Did finance staff verify the last three payment-change requests by phone? Did the help desk refuse the last pretexting call? These are behaviors, and they can be measured.

Why generic content fails

A module written for "all employees" speaks to no one. A hospital's risk lives in shift pressure and shared workstations. A construction firm's risk lives in payment approvals and site devices. A law firm's risk lives in client confidentiality and wire instructions. Training that reflects the actual job, the actual threats and the actual tools lands. Training that does not is forgotten before lunch.

People do not remember training. They remember the moment it was obviously about them.

What works

  • Role-based and short. Ten relevant minutes for a specific role beats an hour of generic content for everyone.
  • Repeated, not annual. Attackers do not wait for the next compliance cycle. Brief, regular touchpoints keep awareness current.
  • Tested. Phishing simulations and tabletop exercises reveal whether the training changed anything — and where to focus next.
  • Extended to leadership. Executives are the most targeted and least trained group in most organisations. Board and leadership sessions close that gap.
  • Documented for the right reasons. Evidence for auditors and insurers is a by-product of a real program, not its purpose.

The new frontier: AI literacy

Staff across every industry are already using generative AI tools, with or without permission. Training that explains what the organisation allows, what data must never be entered, and how to use these tools well is now as important as phishing awareness — and far rarer. It belongs in the same program.

We deliver security and AI-literacy training directly to organisations and through the Rotary IT Professionals Forum Fellowship, always built around the audience's actual work. If your completion rate is high and your confidence is not, that is the program we would rebuild.

Rakesh Babuji
Rakesh BabujiCo-Founder & CEO, Digiward Solutions · Author of The Collaborative Dawn · 25+ years in technology, security and risk across North America, Africa and Asia.

This article is general information for business readers, not legal, regulatory, medical or professional advice. Regulations change; confirm requirements for your situation with qualified counsel or your regulator. Imagery is original AI-generated artwork created with Canva. Frameworks and standards named are trademarks of their respective owners.

Ready to apply this to your organisation?

Start with a free security or AI readiness call — no obligation, and you leave with a clear next step.

Book your free call
DW
Ask Digiward
Guided assistant — pick a question
Book a free call
Free security readiness callBook now