Why leaving security to IT fails
In many companies, security was quietly added to the IT team's list as the business grew, usually without extra people, budget or authority. The chapter identifies three structural problems that have nothing to do with the team's competence. The first is a conflict of interest: IT is measured on uptime, delivery and cost, while security often asks it to slow down, spend or take systems offline, and the brake usually loses. The second is a gap in authority, because security decisions reach into sales, procurement and other functions. The third is a gap in visibility, because warnings travel slowly through tickets and monthly IT reports the board never reads.
A familiar pattern after incidents is an IT manager who repeatedly flagged untested backups and weak remote access in status reports that nobody senior could judge. Security needs an owner independent enough to raise a hand, senior enough to be heard and supported enough to follow through.
The CISO and the reporting line
The CISO, or head of security, designs the programme, chooses the defences, runs exercises and tells leadership honestly how safe the company is and when it needs money or a decision. A good CISO understands the business, translates technical findings into money, time and reputation, leads a programme of controls, people and suppliers, and escalates bad news early and calmly. Being the most technical person in the room is not on the list; if a CISO cannot explain the top three risks to a non-technical board in ten minutes, that gap needs attention.
Each reporting line has trade-offs. Reporting to the head of IT keeps security close to the systems but limits independence. Reporting to the CEO gives standing but may yield little attention. Reporting to the chief risk officer, general counsel or CFO gives independence but distance from the technology. More important than the box on the chart are four conditions: direct access to the board or a committee, the ability to disagree with one's own manager without career damage, a security budget visible as its own line, and a seat when major business decisions are made. A dotted line to the audit committee chair is a common safeguard. At Meridian Foods, the audit committee chair learns that the person named as responsible for cyber security on the insurance renewal has never addressed the committee, and cyber becomes a standing agenda item.
Board and management: owner and manager
The board sets risk appetite, makes sure there is an owner and a plan, oversees and challenges reports, approves big decisions such as major investments, risk acceptances beyond appetite and the company's stance on difficult crisis choices, and prepares for its own role on the worst day. Management runs the programme, identifies and assesses risks, reports honestly including bad news and uncertainty, and leads the response. A simple rule: the board asks whether the company is doing the right things and how it knows; management answers with what it is doing, the evidence and what worries it.
Directors do not need technical expertise to oversee cyber risk, just as non-accountants oversee financial risk through good questions, clear reports and independent audit. A board that spends forty minutes choosing an antivirus product is acting as a manager; one that spends fifteen minutes asking whether the top risks have moved is acting as an owner.
Committees, RACI and the three lines of defence
Board committees go deeper than the full board has time for. They work best when cyber is written into their terms of reference, the CISO presents directly and has a private session without other executives at least once a year, cyber has enough agenda time, and a summary goes back to the full board. Inside the company, a RACI chart records who is responsible for each activity, who is accountable (exactly one person), who is consulted and who is informed. It shows that not everything belongs to the CISO: payment approval sits with finance and regulator notification with legal. The blank cells in a first draft often reveal more risk than a technical audit.
The three lines of defence separate the business and IT teams that own and operate controls, the security and risk functions that set standards and challenge, and internal audit, which tests both independently. The model fails in predictable ways: the first line leaves security to the second, a small security team ends up doing the work it should oversee, or auditors check that policies exist rather than that they are followed. Smaller firms can apply the principle lightly, for example through an annual independent review by an outside firm reported to the audit committee.
Making bad news travel up fast
Every structure in the chapter exists so that bad news reaches people who can act. Good news rises quickly; bad news rises slowly and softens on the way. The chapter recommends explicit escalation rules (for example, a suspected breach of customer data to the CEO and general counsel within two hours), rewarding early warnings rather than only fixes, a bypass route to the board, asking directly what people are most worried about but have not yet said, and reacting calmly to messengers.
At Meridian, after a head of security is appointed with a dotted line to the committee chair, a junior analyst notices a vague notice from a supplier that handles customer emails late on a Friday, checks the escalation rules and calls it in. Within hours the CEO has a short summary of what is known, what is not and when the next update will come. The issue proves limited, and the CEO thanks the analyst by name, which makes it more likely she will do the same next time.
KEY POINTS
Key points
- Name, in writing, the person accountable for cyber risk across the company.
- Give the security lead direct access to the board or a committee, including at least one private session a year.
- Show the security budget as its own line rather than inside IT spending.
- Build a RACI for the ten most important security activities with exactly one accountable person for each.
- Write and test escalation rules that say what must reach the CEO and the board, and how fast.
- Thank people publicly for early warnings, even when the concern turns out to be nothing.
QUESTIONS
Common questions
- Who should be responsible for cybersecurity in a company?
- Security needs a named owner who is independent enough to raise concerns, senior enough to be heard and supported with budget and authority. Leaving it to the IT team by default creates a conflict of interest, a gap in authority and a gap in visibility, however capable that team is. Overall accountability should be written down and visible to the board.
- Who should the CISO report to?
- There is no single correct answer. Reporting to the head of IT keeps security close to systems but reduces independence, reporting to the CEO gives standing but limited time, and reporting to risk, legal or finance gives independence but distance from technology. Whatever the line, the CISO needs direct access to the board or a committee, a safe way to disagree with their manager, a visible budget and a seat in major business decisions.
- What is the board's role in cybersecurity oversight?
- The board sets risk appetite, makes sure there is a capable owner and a plan, challenges reports, approves major decisions and prepares for its own role in a serious incident. It does not choose products or configure controls. Directors can oversee cyber risk without technical expertise by asking good questions, reading clear reports and relying on independent assurance.
- What is a RACI chart for cybersecurity?
- A RACI chart lists, for each security activity, who is Responsible for doing the work, who is Accountable for the outcome (exactly one person), who is Consulted beforehand and who is Informed afterwards. It stops tasks falling between teams that each assume the other is handling them. Blank cells in a first draft often reveal more risk than a technical audit.
- What are the three lines of defence in cybersecurity?
- The first line is the business and IT teams that own systems and operate controls; the second is the security and risk functions that set standards and challenge; the third is internal audit, which independently tests the other two. The model clarifies who does, who checks and who checks the checkers, but it fails when the first line leaves security to the second or when auditors only test whether policies exist.
- How do you make sure bad news about cyber risk reaches the board quickly?
- Write explicit escalation rules that say which events must reach which level and how fast, give the security lead a route to the board that does not depend on their manager, and publicly thank people who raise concerns early. Leaders should also ask directly what people are worried about but have not yet reported, and react calmly to bad news, because everyone learns from how the first messenger is treated.