Principles before layout
The template is a starting point to argue with, not a finished product. What matters most are the principles behind it. Keep it to one page every time, so the security team has to choose what matters and the board can compare quarters at a glance. Show trends rather than snapshots: each measure should carry its current value, its previous value and its target. Write every row in business language. Treat colours with suspicion, and ask for a short written comment beside each one. Balance lagging indicators, such as incidents, with leading ones, such as how quickly weaknesses are fixed.
The five blocks
The page reads like a short letter followed by a small table. It opens with a headline of two or three plain sentences from the head of security on the overall picture. Next come the three to five top risks, each with an owner and a direction arrow, matching the enterprise risk register. The third block is a table of eight to ten key indicators. The fourth lists notable events since the last meeting, and the fifth states what the security team needs from the board, or says plainly that nothing is needed.
Suggested indicators include multi-factor authentication coverage, time to fix critical weaknesses, backup restore tests, critical supplier reviews, phishing reporting, incidents and near misses, time to detect and contain, unsupported systems, policy exceptions and roadmap delivery. The test for any row is whether a director reading it could ask a sensible question.
Honest colours and a quarterly routine
Colours should follow thresholds agreed with the board in advance, not the mood of whoever prepares the report. The book warns of two traps: the watermelon, green outside and red inside, such as backups that are complete but never restored; and permanent amber, where a row stuck at amber for four quarters is really red.
The page is drafted about three weeks before the meeting, reviewed with an executive sponsor for clarity rather than softening, and sent out a week ahead. In the meeting, assume it has been read and spend the time on the headline, red rows, changing trends and decisions.
The example from Meridian Foods
Meridian's first draft ran to four pages and forty measures until the chair asked which ten numbers would survive on a single page. The revised page flagged that the warehouse system had never been restored from backup. The board approved a weekend test, which took nineteen hours instead of the expected four. Finding that out on a quiet weekend rather than during an attack showed the real purpose of the dashboard: to start the right conversations, not to prove that all is well.
KEY POINTS
Key points
- Limit board cyber reporting to one page and keep the same shape every quarter.
- Show each measure with its current value, previous value and target.
- Agree colour thresholds with the board once, in advance.
- Treat any row that has been amber for four quarters as red and ask what is blocking progress.
- Spend meeting time on the headline, red rows, changed trends and decisions, not on reading every row.
- Mark missing data as not yet measured instead of leaving the row off.
QUESTIONS
Common questions
- What should be on a board cybersecurity dashboard?
- The Boardroom Firewall suggests one page with five blocks: a short headline from the head of security, the top three to five cyber risks with owners and trend arrows, a table of eight to ten key indicators, notable events since the last meeting, and the decisions the board is being asked to make. Each indicator should show the current value, the previous value and the target.
- Which cyber security metrics should a board track?
- Useful board-level indicators include multi-factor authentication coverage, time to fix critical weaknesses, results of backup restore tests, critical supplier reviews, phishing reporting rates, incidents and near misses, time to detect and contain, unsupported systems, policy exceptions and progress on the security roadmap. Not every company needs every row; the test is whether a director could ask a sensible question from it.
- How should red, amber and green ratings be set on a cyber dashboard?
- Colours should come from thresholds agreed with the board in advance, not from the judgement of the person writing the report. For example, multi-factor coverage might be green above 98 per cent, amber from 90 to 98 and red below 90. Each colour should carry a short comment explaining what the colour does not show.
- What is a watermelon metric in security reporting?
- A watermelon metric is green on the outside and red on the inside: the measure is technically met, but the reality underneath is poor. A typical case is backups reported as fully complete when nobody has ever tried to restore one. A written comment beside each colour is the main defence.
- What should we do if we have no data for a dashboard indicator?
- Say so openly rather than dropping the row. A row marked as not yet measured, with a date when data will be available, is more useful than a quiet omission. The gap is itself a finding, because it shows the board where the security team's visibility ends.