Start by finding out what is true
Every organisation tells a story about its own security, and it is usually a little rosier than reality, not through dishonesty but because people grow used to the gaps. A new leader's first task is to hear that story and test it gently. In the first two or three weeks, meet the security lead, the head of IT, the general counsel, the CFO, the head of operations and the owners of the largest supplier relationships, and ask each the same questions: what worries you most, what would happen if our main systems stopped for a week, when did we last practise for that, and what do you need that you lack? Where the answers differ, the gaps usually lie.
Then gather a handful of documents: cyber entries in the risk register, the latest security report, the incident response plan, the insurance policy and application, any recent external assessment and the list of critical suppliers. The point is not technical detail but whether they exist, are current and connect to each other. In the book, Meridian Foods' new chief executive, Claire Morgan, kept a notebook of surprises: an outdated response plan naming departed staff, backup tests less frequent than the insurance form claimed, and a weakened reporting line for the security lead. She brought the notebook, not a plan, to the chair and asked for ninety days.
A rough baseline against the six NIST CSF 2.0 functions (Govern, Identify, Protect, Detect, Respond, Recover), rated simply as strong, adequate or weak, gives a shared starting point. Many organisations are stronger at Protect, where tools are bought, than at Govern, Respond and Recover.
The first ninety days in three phases
Days 1 to 30 are for listening and mapping, plus one or two quick, visible decisions that cost little and fix something real, such as confirming who can declare an incident or enforcing multi-factor authentication on every executive account, starting with your own. Days 31 to 60 are for the basics: multi-factor authentication, separated and tested backups, timely patching, staff awareness and logging, each gap with an owner and a date. This is also the time to give the security lead a direct route to the chief executive and the committee chair. Days 61 to 90 are for rhythm: the first executive tabletop exercise, an agreed format for board reporting, and a short paper on what was found, done and proposed.
The realistic goal at ninety days is not "the company is secure" but shared understanding, the worst gaps closed or on track, a current and rehearsed response plan, honest regular reporting and a credible roadmap for the next twelve to eighteen months. Common traps are the shiny purchase made before the basics, the external report nobody acts on, the reorganisation that consumes the quarter, and going quiet after the board paper.
Maturity in three stages
The book describes three stages. At the basics stage, the essential controls are in place and consistently working; many companies hit by serious incidents had not finished this stage. At the managed stage, security is part of how the company runs: risks have owners, suppliers are assessed before contracts, new projects are reviewed early, tabletop exercises happen at least yearly, and processes work even when people change. At the resilient stage, the company can absorb a serious incident and recover quickly, with mapped critical services, tested recovery times and the ability to operate in a degraded mode.
Not every area needs to be resilient; the right stage depends on risk appetite. Many organisations stall at the managed stage because further progress needs business-wide investment, so moving to resilience is usually a business programme sponsored by the CEO or COO rather than a security project.
The roadmap, and what leaders should own
A roadmap of twelve to thirty-six months should start from the top risks in the register, sequence work sensibly, respect the small team's real capacity and report progress in business terms. The guiding rule is to delegate the doing and keep the deciding. Leaders own the risk appetite, priorities, resources, the reporting line, culture and crisis decisions. When Meridian's security team asked for an expensive threat intelligence platform, Claire asked which risk it reduced, where that risk stood against appetite and what the purchase would delay. Unfinished restore tests and unassessed suppliers mattered more, and the request was deferred.
Keeping it going, and leading with steadiness
Cyber risk drifts to the margins because success looks like nothing happening. The cure is structure rather than anxiety: fixed points on the board calendar, tabletop exercises scheduled a year ahead with rotating scenarios, a roadmap tied to the budget cycle, the insurance renewal used as an outside check, cyber objectives for the relevant executives, and an annual repeat of the baseline. Security is a garden that needs steady care, and the programme should survive the departure of any individual, including the leader.
The book closes on steadiness. Fear leads to overspending, blame, silence and exhaustion. A steady leader takes the risk seriously, asks clear questions, invests in the basics, prepares in calm weather and thanks people for raising concerns.
KEY POINTS
Key points
- Spend the first weeks listening to key people and comparing where their answers differ.
- Make one or two quick, visible fixes, such as MFA on every executive account starting with your own.
- Rate the company roughly against the six NIST CSF 2.0 functions as a shared starting point.
- Ask of any security purchase which risk it reduces and what it would delay on the roadmap.
- Put cyber reviews, exercises and the insurance renewal on the calendar a year in advance.
- Repeat the baseline assessment each year and update the roadmap as the business changes.
QUESTIONS
Common questions
- What should a new CEO do first about cyber security?
- Start by listening rather than buying. In the first few weeks, meet the security lead, IT, legal, finance, operations and owners of key supplier relationships, ask each the same questions about their worries and preparedness, and gather a few core documents such as the incident response plan and insurance application. Differences between people's answers usually reveal the real gaps.
- What does a 90-day cyber security plan for leaders look like?
- The book divides it into three phases of about a month each. Days 1 to 30 are for listening and mapping plus one or two quick, visible fixes. Days 31 to 60 are for closing gaps in the basics, such as multi-factor authentication, tested backups and patching, and fixing the security reporting line. Days 61 to 90 are for running a first executive tabletop exercise, agreeing board reporting and presenting a short paper and roadmap.
- What are the stages of cyber security maturity?
- A simple model has three stages. Basics means essential controls such as multi-factor authentication, separated and tested backups and timely patching work consistently. Managed means security is built into how the company runs, with owners, supplier assessments, regular reporting and yearly exercises. Resilient means the company can absorb a serious incident and recover quickly because it has mapped critical services and tested recovery.
- Which parts of cyber security should a CEO own personally?
- A useful rule is to delegate the doing and keep the deciding. Specialists should design controls, choose tools and run investigations. Senior leaders should own the risk appetite, the priorities, the resources given to security, the reporting line for bad news, the culture, and the business trade-offs that only an executive can make during a crisis.
- How do we keep cyber security on the agenda after the initial push?
- Rely on structure rather than anxiety. Put cyber reviews at fixed points on the board calendar, schedule tabletop exercises a year ahead with rotating scenarios, tie the security roadmap to the budget cycle, use the insurance renewal as an annual outside check, include cyber objectives in relevant executives' goals, and repeat the baseline assessment every year.