Three phases, taken in order
The plan is written at governance level; configuring firewalls and rebuilding servers is left to the technical team. It has three phases of about a month each: listen and map, fix the basics and the reporting line, and set the rhythm. The book asks readers not to run them in parallel, because the pacing is the point. Each phase sets out a goal, the people to involve, a short weekly checklist and a description of what done looks like.
Days 1 to 30: listen and map
The aim is to find out what is true before changing anything. In the first week the leader meets the security lead, head of IT, general counsel, finance, operations and the owner of key supplier relationships one to one, asking each the same few questions, such as what would happen if the main systems stopped for a week. Over the next weeks they gather a small set of documents, from the risk register entries to the insurance application, and build a rough baseline against the six functions of NIST CSF 2.0. One or two quick, visible decisions follow, for example confirming who can take systems offline or enforcing multi-factor authentication on executive accounts, starting with the leader's own. The month ends with a single page of questions that could not be answered clearly.
Days 31 to 60: fix the basics and the reporting line
The second month closes the most dangerous basic gaps and makes sure bad news can travel quickly. The security lead assesses multi-factor authentication, separated and tested backups, timely patching, staff awareness and logging as in place, partial or missing. Every gap receives an owner and a date, even if the fix takes longer than a month. If the senior security person sits several layers down, a direct route to the leader and the audit or risk committee chair is created. The leader also checks that the controls declared on the cyber insurance application really exist, and that the riskiest critical suppliers are being assessed.
Days 61 to 90: set the rhythm
The final month turns a burst of attention into routine. It includes a first executive tabletop exercise, agreement on a one-page board dashboard, and a calendar of quarterly committee reviews, an annual board deep dive, a yearly tabletop and the insurance renewal. It ends with a short ninety-day paper to the board covering what was found, what was fixed, what remains and what is needed over the next twelve to eighteen months, with the next review date agreed before the meeting closes.
The book adds a few habits to make the plan stick: give every item an owner, keep the paper to about five pages, schedule the recurring rhythm before attention fades, and repeat the baseline yearly. In the book's words, cyber security is a garden, not a monument.
KEY POINTS
Key points
- Spend the first month finding out what is true before changing anything.
- Make one or two visible early decisions, such as enforcing MFA on your own account.
- Give every basic security gap a named owner and a date.
- Create a direct reporting route from the security lead to you and the audit or risk committee chair.
- Run an executive tabletop exercise and put the recurring review calendar in place by day 90.
- Repeat the baseline assessment every year.
QUESTIONS
Common questions
- What should a new CEO do about cybersecurity in the first 90 days?
- The Boardroom Firewall proposes three phases of about a month each. Days 1 to 30 are for listening and mapping the real situation; days 31 to 60 for closing the most dangerous basic gaps and fixing the reporting line; and days 61 to 90 for setting a lasting rhythm, including a tabletop exercise and a ninety-day paper to the board. The phases should be taken in order rather than in parallel.
- What documents should a new board member ask for on cyber risk?
- A useful starting set is the cyber entries in the risk register, the latest board or committee security report, the incident response plan, the cyber insurance policy and last application form, any recent external assessment, and a list of the most critical suppliers. The point is to check that each exists, is current and is consistent with the others.
- What quick cyber security wins can a new leader make?
- In the first month, the book suggests one or two low-cost, visible decisions that fix something real. Examples are confirming who has authority to declare an incident and take systems offline, updating the contact list in the incident response plan, or enforcing multi-factor authentication on executive accounts, starting with the leader's own.
- Which basic security controls should leadership check first?
- The plan asks the security lead for an honest status on multi-factor authentication, separated and tested backups, timely patching, staff awareness and good logging, each marked as in place, partial or missing. Every gap should then be given a named owner and a date, even if some fixes take longer than a month.
- What should a 90-day cyber report to the board contain?
- It should be short, around five pages at most, and structured around what was found, what has been fixed, what remains, and what support is needed over the next twelve to eighteen months. The board should agree the date of the next review before the meeting ends.