OneNym.idDiscuss deployment
English

THE BOARDROOM FIREWALL / Chapter 11

People Are the Perimeter: Culture

Locks and alarms are only half of a building's safety; the other half is what the people inside do when nobody is checking. This chapter argues that people are a company's perimeter, not its weakest link, and that leaders shape that perimeter far more than technology does.

This page summarises the chapter. The full text is in the book.

Why culture beats tools

Most security incidents start with a person: a click on a convincing email, a payment sent because the request seemed to come from the boss, an oddity noticed and never mentioned. Industry studies consistently find a human element in a large majority of breaches. Technology such as email filtering and multi-factor authentication helps and deserves funding, but it cannot stop a clerk following a persuasive phone call or make a tired analyst look properly at an alert.

The book defines culture plainly as what people do when nobody is checking. In a strong culture, people verify unusual requests without embarrassment and report mistakes quickly, the way good cooks treat hygiene as part of cooking rather than an obstacle. Culture adapts to new threats before tools exist for them, and its most effective measures cost very little. It also changes slowly, and mostly through what leaders do rather than what they say. A test at the fictional Meridian Foods, where unfamiliar USB sticks labelled as salary documents were left in the car park and several were plugged in within a day, showed that curiosity and helpfulness are simply human.

Tone from the top

Staff watch what leaders praise, skip and make time for. A chief executive who holds the secure door for a stranger, or a finance director who shares his password with an assistant just this once, teaches everyone that rules bend for important people, which is exactly what impersonators exploit. Small visible acts work the other way: asking a visitor to badge in, mentioning that you reported a suspicious email, completing training on time.

When Meridian's phishing simulation included senior leaders for the first time, the chief executive clicked. He asked for the result to be included in the company-wide report and told the next all-hands meeting what had happened, explaining that such emails are designed to catch people when they are busy. Reporting of simulated emails rose sharply the next quarter. Tone from the top also means giving security proper time on the agenda and describing the security team as a partner rather than a department of no.

Reporting without fear

The single most valuable habit is early reporting. A phishing email reported within minutes can be removed from every inbox, and a mistaken payment reported within the hour can sometimes be recalled. People report early only if they expect fair treatment. Blame-free reporting means honest mistakes reported promptly are treated as learning, while deliberate wrongdoing, repeated recklessness and concealment can still carry consequences. Aviation's confidential near-miss reporting is the model.

The book sets out four steps: make reporting easy through one obvious button and one contact point; respond quickly and warmly every time; thank reporters publicly, as Meridian does in a monthly good catch note; and protect an honest reporter even when a mistake causes real cost. When a Meridian logistics coordinator emailed delivery addresses to the wrong supplier contact, she called the security line instead of hoping nobody would notice; within an hour the recipient had confirmed deletion and the privacy team had recorded the event.

Training that changes behaviour

An annual forty-five-minute online course satisfies auditors but rarely changes what people do on a busy afternoon. Training that works is short and frequent, specific to each role, built on realistic examples from the company's own experience, respectful rather than fear-driven, and taken by leaders too. Simulated phishing helps only if staff see it as fair practice rather than a trap.

Measurement matters. Completion rates show compliance, not behaviour, and belong in the audit file rather than on the board dashboard. Click rates depend heavily on how difficult the simulation was. The book prefers the report rate and the time to first report, along with behavioural measures such as the share of bank-detail changes verified by call-back.

Incentives, the easy path and the leader's own behaviour

Targets can quietly work against security. At Meridian, a fraudster persuaded a customer-service agent to change the delivery address on a large standing order, skipping two verification questions. The team's dashboard showed call handling time in large green numbers and nothing about verification. After the dashboard and monthly recognition were changed to reward verification and stopped fraud, call times rose slightly and fraud through the desk fell to almost nothing over six months. The same logic applies more widely: make the secure way the easy way, through tools such as password managers, single sign-on and a one-click report button.

The loudest signal is the leader's own behaviour: accepting the same controls as everyone else, thanking staff who call back to verify a genuine request, asking how something happened rather than who did it, keeping security on the calendar outside crises, and supporting whoever brings bad news.

KEY POINTS

Key points

  • Accept the same security controls as every other employee, with no personal exceptions.
  • Thank, in person and in public, anyone who calls you back to verify a request.
  • Provide one obvious reporting button and one contact point, and reply to every report.
  • Ask what happened to the last three people who reported their own mistakes.
  • Replace course-completion figures on the board dashboard with report rates and time to first report.
  • Review performance targets for any that reward skipping security steps.

QUESTIONS

Common questions

How do you build a strong security culture in a company?
Culture is shaped mainly by what leaders do. Leaders should accept the same security controls as everyone else, visibly thank people who verify requests or report problems, and make the secure way of working the easy way. Combined with an easy reporting channel and short, role-specific training, these habits build a workforce that verifies unusual requests and reports mistakes quickly.
What is blame-free reporting in cybersecurity?
Blame-free reporting means that honest mistakes reported promptly are treated as a chance to learn, not as disciplinary matters. Deliberate wrongdoing, repeated recklessness after coaching, and concealment can still carry consequences. The aim is that people report problems such as a mistaken email or a clicked link within minutes, while the damage can still be limited.
Does security awareness training actually work?
An annual online course mostly satisfies auditors and rarely changes behaviour. Training that works is short and frequent, specific to each role, based on realistic examples, respectful in tone, and taken by leaders as well. Its effect should be judged by behaviour, such as how many people report suspicious emails and how quickly, rather than by completion rates.
What is a better metric than phishing click rate?
The report rate, the share of recipients who report a simulated or real phishing email, is generally a better indicator of culture because it measures active defence. The time from delivery to the first report shows how quickly the organisation can respond. Click rates still have a place, but they depend heavily on how difficult the simulated email was.
Can performance targets undermine security?
Yes. A team measured only on speed, such as call handling time in customer service or feature delivery in engineering, will tend to skip verification or postpone security reviews under pressure. Leaders should look for these hidden conflicts and measure the secure behaviour alongside the speed target, for example verification compliance or fraud attempts stopped.