Frequently Asked Questions about Hacktivism and Cyber Resilience
Hacktivism raises questions that sit awkwardly between technology, politics, public trust and the everyday business of getting things done. These pages turn the article’s central arguments into practical answers for leaders, service owners, trustees and anyone expected to make a sensible decision while the available information is incomplete. You do not need to be a security specialist, but you do need to ask what the disruption means for the people depending on your organisation.
The answers distinguish reported evidence from my interpretation and recommendations. They explain essential terminology, identify useful questions for suppliers and colleagues, and suggest ways to test whether an assurance survives contact with reality. The answers are grouped by topic, from understanding the threat to testing resilience. Read them in sequence or go directly to the problem in front of you. A small charity and a large infrastructure operator will need different arrangements, even where the governing principles are similar.
Treat this as a basis for informed discussion and proportionate action, not a universal incident manual or legal checklist. Adapt the advice to your services, resources, jurisdiction and exposure. The linked sources were consulted on 15 November 2026. Where safety, sensitive information or legal obligations are involved, bring in the relevant specialists early. Knowing when to ask for help is a professional skill, not an admission that your job title was printed in error.
Understanding the Threat and Assessing Claims
In this article, hacktivism means unauthorised digital activity undertaken, or presented, in support of an ideological cause. The wording is deliberately cautious. It distinguishes the act from the explanation offered for it. A group’s political language may help explain its targeting or publicity, but it does not establish who its members are, whether somebody supports them or what they actually achieved.
For a leadership team, the useful starting point is consequence. Has a service become unavailable? Is information exposed or altered? Are employees or customers being threatened? Could a physical process have been affected? These are different problems, requiring different people and decisions. Treating them all as “a hacktivist incident” risks concealing the very distinctions that make a response useful.
I would ask the incident lead to prepare a short account separating observed facts, the attacker’s assertions and the questions still being investigated. Name the affected service, explain its purpose and describe the practical effect on users. Include the source and time of each important observation. That gives colleagues something they can challenge and update, rather than a label they are expected to accept.
Keep motive and method separate throughout. An organisation can defend itself against interference without deciding that everyone who shares the underlying grievance is an adversary. Equally, sympathy with a cause is not a reason to overlook harm. My preference is to investigate the conduct, protect those affected and reserve judgement where evidence is missing. We can debate the terminology once the service works. A political slogan, however elegantly designed, is not an incident assessment.
Take threats seriously enough to prepare, but not so dramatically that the attacker ends up writing your investment plan. The NCSC’s January 2026 warning says technically simple denial-of-service attacks can significantly disrupt organisations and prevent people accessing essential services. The important distinction is between how difficult an attack is to perform and what happens when the target cannot withstand it.
I would assess a threat against the services named, the organisation’s exposure and any evidence of activity. A vague boast, a credible warning about a priority service and confirmed interference with a control system should not trigger identical responses. Set escalation thresholds beforehand, including what would bring in privacy, safeguarding or operational-safety specialists. Record the reasoning so that a later change of position can be explained.
Ask service owners to describe harm in terms a customer would recognise. Can someone obtain urgent information, complete a necessary application or reach support? How long could the interruption continue before consequences become materially worse? Consider the staff needed to handle alternatives, investigate the event and communicate with users. Preparedness should address that workload, not simply the capacity of the website.
Finally, make uncertainty part of the briefing rather than a reason to stop thinking. State what is known, what is plausible and which evidence would change the assessment. Review the position at agreed intervals and after significant developments. My view is that calm preparation is more useful than either dismissal or catastrophe theatre. A service can deserve urgent protection without its attacker deserving a starring role in the evening news.
Treat an attacker’s claim as information to assess, not a conclusion to inherit. A screenshot, countdown or declaration may warrant investigation without proving present access or operational effect. Ask exactly what is claimed: an unavailable website, possession of information, altered settings or control of a physical process. Those assertions need different checks and may justify different protective actions while verification continues.
Compare the claim’s timing and scope with the organisation’s own observations, provider evidence and reports from users. Ask whether the material concerns the actual service and whether it establishes anything beyond what is already public. Competent investigators should assess technical material. Staff should not download suspect files or engage hostile actors simply to satisfy a manager’s curiosity. Define the question first and obtain evidence through an appropriate, safe process.
Use a briefing structure that makes uncertainty visible: confirmed facts, attacker claims, unresolved questions and next actions. For each significant claim, record its source, the assessment so far and what would change that assessment. Avoid turning repeated publication of the same assertion into apparent corroboration. Equally, do not dismiss everything because the source has an obvious incentive to exaggerate. Untrustworthy people can still make claims requiring urgent investigation.
Decide separately what action is justified before verification is complete. Supporting a threatened employee or preparing an alternative service route may be sensible without endorsing the attacker’s narrative. A public accusation or assurance will need its own evidential basis. Update the assessment when new information arrives, including when that means correcting an earlier view. My recommendation is disciplined scepticism rather than reflexive disbelief. The briefing should help people decide what to do next, not become a fog machine producing impressive uncertainty at regular intervals.
No. Shared political objectives, encouragement, resources and operational direction are different relationships. “State-aligned” should not silently become “under government command”. The NCSC’s January 2026 warning describes groups within its scope as operating outside direct state control. The international critical-infrastructure advisory describes a varied picture, including assessed direct or indirect support for some groups. Preserve the scope of those assessments rather than combining them into a universal verdict.
When a briefing uses a relationship label, ask what it means in that particular source. Is the assessment about ideological sympathy, financial or technical assistance, an organisational link or direction of a specific operation? Who made the assessment, when, and with what confidence? The answer may remain incomplete. That is a reason to use careful language, not an invitation to fill the gap with whichever geopolitical theory feels most convincing.
For practical response, separate attribution from protective decisions. Restoring an essential service, preserving evidence or involving operational-safety expertise does not normally require a final account of who directed the activity. Publicly alleging state responsibility is a different decision with potentially wider consequences. Use the relevant specialist and official channels rather than allowing an incident call to improvise foreign policy between supplier updates.
Maintain a record of the evidence supporting the organisation’s position and review it when credible new reporting appears. Avoid describing an assessment as certainty or treating a lack of public evidence as proof that no relationship exists. My rule is to name the source, preserve its qualifications and state what remains unknown. Conviction about defending customers and staff can coexist with caution about an attacker’s sponsors. We should be decisive about justified action and appropriately modest about matters the available evidence cannot settle.
Europol’s July 2025 account of Operation Eastwood reports disruption of infrastructure comprising more than 100 computer systems and seven arrest warrants. It also describes participants using automated tools, motivated by ideology and rewards. These are specific reported results and observations. They should not be stretched into proof that every participant was highly capable, that every attack succeeded or that the network was permanently eliminated.
The first management lesson I draw is that international cooperation can impose tangible costs. Organisational evidence and coordination with appropriate authorities therefore deserve attention alongside technical defence. Establish who can handle those contacts, what information may be shared and how relevant records will be preserved. Do that through responsible channels, with appropriate advice, rather than encouraging individual employees to conduct their own cross-border investigation.
The second lesson concerns persistence. A campaign using shared tools and participation can create sustained work even where individual attempts are contained. Plan provider coordination, decision cover and staff rotation for disruption that lasts beyond the first emergency call. Agree thresholds for reviewing additional expenditure or changing continuity arrangements. Protecting the response team’s capacity is part of protecting the service, not an optional wellbeing exercise after the operational work.
Finally, distinguish an enforcement announcement from an organisational closure decision. Welcome a successful operation, assess relevant new information and retain the capability needed for the services you provide. Review whether the case exposes assumptions in your own protection or escalation arrangements, without assuming the same actors or methods will target you. My interpretation is limited but useful: cooperation matters, accessible tools can aggregate pressure, and resilience remains your responsibility. A welcome result is not a lifetime guarantee, however much everyone would appreciate a quieter week.
Separate observed capability, informed forecasts and unsupported claims. The article discusses assessments about AI’s evolving role in malicious activity. That does not establish that a particular attacker has autonomous capabilities, or that an individual incident used AI. Before changing priorities, ask what the evidence actually shows and whether the proposed conclusion belongs to the source or to somebody interpreting it. Preserve that distinction in briefings.
Start with the defensive decision. Does the claim change which service needs protection, the resources required for investigation or the way evidence should be checked? If the answer is unclear, more dramatic language will not make the requirement clearer. Compare the proposed response with known gaps in ownership, access, supplier protection and continuity. A fashionable threat should not automatically displace a demonstrated weakness simply because it sounds more modern.
Apply the same discipline to AI used inside the response. I would require human verification of consequential summaries, source references and recommended actions before they enter a leadership briefing. Make somebody accountable for checking what the underlying evidence supports, including uncertainty and conflicting information. Follow the organisation’s approved arrangements for handling sensitive material. Useful assistance is welcome, but a fluent paragraph should not acquire decision-making authority merely by sounding confident.
Review assumptions when credible evidence changes, rather than turning enthusiasm or scepticism into a permanent identity. Ask specialists to explain developments in terms relevant to your services and record why a change in defensive investment is justified. Exercise the possibility of misleading or incorrect information without assuming every convincing claim is AI-generated. My view is that the right combination is curiosity, verification and ordinary operational discipline. We can prepare for changing capabilities without treating forecasts as facts or outsourcing the responsibility to think.
Protecting Services and Working with Providers
Protect the outcomes whose loss would cause the greatest human or operational harm, not automatically the systems with the highest profile. I would start with safety, essential services, sensitive information, urgent customer needs and important obligations. Ask what people would be unable to do, who would be affected and whether any group has particularly limited alternatives. That makes prioritisation a service decision rather than a technology popularity contest.
Bring service owners, operations, customer support and technology colleagues together to trace the route from a user’s need to a completed outcome. Include sign-in, payments where relevant, communications, suppliers and the people who make decisions along the way. Do not assume that apparently separate services are independent. Challenge whether their alternatives rely on the same provider, credentials, connection or unavailable information.
For each priority outcome, agree a tolerable interruption, a named owner and an alternative route. Explain what the fallback can and cannot deliver. If manual processing is proposed, identify who is trained, how urgent cases will be prioritised and how records will be reconciled afterwards. These are recommendations for practical planning, not a suggestion that every service needs an elaborate duplicate system.
Then test a complete user journey under realistic restrictions. Ask someone to demonstrate how a customer would obtain the outcome when the main website and sign-in are unavailable. Include accessibility and people unable to switch channels easily. Record failed assumptions and fund the most consequential gaps first. A resilient homepage is not much comfort if the booking system behind it has gone for a lie-down. The priority is the person trying to get something done.
Begin with the service you need to preserve, rather than the provider’s headline capacity. Ask whether protection covers the whole customer journey, including sign-in, search and other demanding application functions. Clarify which network, protocol and application attacks are addressed. Those terms refer to different points at which a service can be overwhelmed, so an impressive answer about one does not automatically answer the others.
Then explore activation and limits. Does protection operate continuously or require a request? Who makes that request, how is it authenticated and what happens when the nominated contact is unavailable? What happens above contracted limits, which additional charges might arise and who can authorise them? Ask which changes the provider can make independently and which require your organisation’s approval. Record the responsibilities in language both teams understand.
Discuss evidence and recovery as well as traffic handling. What information will the provider share, how quickly, and through which contact? How will you distinguish an attack from another service problem? Agree out-of-hours escalation, arrangements for prolonged disruption and how protective measures will be reviewed or withdrawn. Include relevant upstream dependencies instead of assuming your immediate supplier controls every part of the service.
Finally, test the arrangement using an agreed, safe exercise. Confirm that named contacts respond, authorisation works and both sides can coordinate without relying on an unavailable communications channel. Do not conduct unapproved stress tests against live services. Review the result with the service owner, procurement and technical teams, then close identified gaps. “Included in the contract” is a useful beginning, not a successful demonstration. The awkward commercial conversation belongs before the incident, preferably while everyone has had some sleep.
A proportionate approach begins with knowing what must keep working. For a small organisation, I would identify a short list of priority services, the people responsible for them and the providers on which they depend. Describe the consequences of losing each service, including urgent customer or beneficiary needs. You do not need a large security department to decide which interruption would create the greatest difficulty.
Concentrate effort on understandable commitments and usable arrangements. Confirm provider protection and support, restrict privileged access and keep reliable contact details somewhere accessible during disruption. Decide who can approve urgent expenditure and who takes over if the usual decision-maker is away. Where specialist help is not available internally, agree how competent assistance will be obtained. An emergency is an expensive time to discover nobody knows whom to call.
Develop a modest fallback for the most important outcome. That could involve an independently accessible source of updates, an alternative service route or a prioritised manual process, depending on the work. Treat these as options to assess, not universal prescriptions. Check confidentiality, staffing, accessibility and the information people need to use the alternative. A fallback that depends on the failed sign-in system deserves another look.
Run a short exercise with the people who will actually respond. Use a plausible outage and ask them to demonstrate decisions, contacts and customer support. Record a manageable set of improvements with owners and review dates. Avoid copying a much larger organisation’s paperwork merely because it looks reassuring. My preference is for a plan that fits the organisation’s capacity and works when someone is on leave. A small, tested arrangement beats a comprehensive document that nobody has opened since the audit.
Begin with the people who have least room to absorb disruption. For a council, planning might focus on urgent statutory services and residents unable to use alternatives. For a charity, priorities might include safeguarding, beneficiary confidentiality, donation journeys or volunteer coordination. These are planning examples, not claims that every organisation faces identical risks. Ask the service owner and frontline staff which outcomes genuinely cannot wait.
Examine alternatives from the user’s perspective. Can someone with limited connectivity, accessibility needs or no private place to telephone still obtain help? What happens when the main website and its sign-in service are unavailable together? Check that published instructions are understandable and reachable without those systems. Do not ask vulnerable people to navigate an organisational chart to discover which department owns their problem.
Plan the human capacity behind the alternative. Identify who answers calls, handles urgent cases, authorises exceptions and coordinates partner organisations. Agree what information staff need, how it will be protected and how decisions will be recorded. If demand exceeds capacity, establish a defensible prioritisation process before the incident. A telephone number is not a complete fallback unless somebody can answer it and deliver useful assistance.
Test a realistic service journey with the relevant operational and safeguarding expertise. Where appropriate, seek feedback from people familiar with users’ needs rather than assuming a technically available channel is accessible in practice. Review shared dependencies with partners and suppliers, and clarify who communicates changes. My judgement is that resilience should be assessed by the essential outcome preserved, not the number of channels advertised. An alternative that works only for confident, well-connected users leaves precisely the people needing most support to manage the largest gap.
Operational technology monitors or controls physical processes, so unauthorised access creates questions about safe operation as well as digital availability. The December 2025 international advisory reports varying impacts, including physical damage, and recommends reducing public internet exposure, mapping assets and access paths, and strengthening authentication. That evidence warrants attention without turning every public website problem into a claim of industrial sabotage.
When credible evidence points to control-system access, involve the responsible engineers and service operators immediately. They need to assess whether settings changed, readings can be trusted and operations can continue safely. Technology specialists contribute important evidence, but should not make physical-process judgements outside their competence. Equally, engineering teams need a clear route to the incident lead and an understandable account of what has been observed.
Before an incident, ask who owns remote maintenance arrangements and which suppliers can connect. Establish how access is authorised, reviewed and suspended, and what operational consequences would follow suspension. Identify the information and expertise needed if normal visibility is lost. Any proposed changes to live environments need appropriate engineering review and safe implementation. Removing exposure is a priority, not permission to disconnect equipment impulsively and hope the process tolerates it.
Exercise the decision-making without creating live operational risk. A suitable scenario might ask how operators would respond to uncertain readings and unavailable remote support, using established safety arrangements. Record who can authorise operational decisions, how evidence is preserved and what conditions would permit a return to normal. My recommendation is to build this into operational governance rather than bolt it onto a website response plan. The controlling question is whether people can trust the process and operate safely, not whether the attacker’s branding looks amateurish.
Responding to Incidents and Supporting People
My suggested starting point is to establish an incident lead and confirm the service effects. Ask what is failing, when it was first observed, which users are affected and which evidence supports the assessment. Bring in the relevant technical teams and providers. Keep a decision log from the outset, including significant observations, actions, owners and times. Memory becomes surprisingly creative after several difficult hours.
Check early for matters that change the response: possible personal-data exposure, threats to individuals, safeguarding concerns or indications of interference with physical operations. Involve the responsible specialists where those indicators exist. This is not an instruction to follow a rigid sequence while a safety problem develops. Immediate danger, credible intrusion evidence or an operational concern may justify a different priority and more urgent escalation.
In parallel, decide whether continuity arrangements should be activated. Confirm that the alternative works before directing customers towards it, and check whether its staff can manage the likely demand. Agree who can authorise urgent expenditure, service suspension or protective changes. Avoid improvised changes outside the team’s competence, particularly in operational environments. The objective is to contain harm without creating an additional problem through hurried guesswork.
Prepare a short internal briefing that separates facts, attacker claims, unanswered questions and next actions. Publish practical service information where appropriate, without guessing at attribution or offering unsupported reassurance. Set the next review time and name the decisions that cannot wait. This first-hour outline is management advice, not a universal technical runbook. Your organisation’s circumstances, legal obligations and incident plan should determine the detailed response. Calm coordination is more valuable than everybody doing something impressive at once.
No. A distributed denial-of-service attack, or DDoS attack, attempts to make a service unavailable by overwhelming it from multiple sources. It addresses availability: can legitimate users obtain the service? Data theft concerns confidentiality: has somebody obtained information they should not possess? An outage alone does not answer that second question. Keeping the distinction clear prevents both unnecessary alarm and unjustified reassurance.
The sensible response is to investigate relevant intrusion indicators while dealing with the interruption. Ask technical teams what evidence they can obtain from the affected service, identity systems and providers. Establish whether there are unexplained changes, unauthorised access indications or credible claims involving non-public information. These are questions for competent investigators, not a request for colleagues to experiment with systems while the incident is developing.
Brief leaders separately on service effects and possible compromise. A useful update might explain that access is disrupted, mitigation is under way and the team has not yet established whether any information is affected. Describe the limits of that statement. “No evidence identified so far” means something different from “We have proved no access occurred”. Avoid compressing a qualified technical assessment into an absolute public promise.
If personal data may be involved, bring privacy specialists into the assessment early and consider applicable notification requirements. If there are threats to individuals, do not postpone support while investigators settle the precise technical mechanism. My recommendation is to maintain linked but distinct workstreams for continuity, investigation and people’s needs. That allows each decision to follow the evidence it requires. A DDoS diagnosis should neither manufacture a data breach nor become a convenient excuse to stop looking for one.
Start by listening and establishing a confidential point of contact. People whose information is exposed, or who receive threats, need practical help rather than a request to remain calm for the organisation’s benefit. Ask about immediate safety and safeguarding concerns, explain what support is available and involve the appropriate specialists. Where there is immediate danger, prioritise the relevant emergency response rather than waiting for the cyber investigation.
Preserve relevant evidence without circulating the exposed information more widely. Agree who needs access, how it will be protected and who will coordinate any appropriate reporting or removal requests. Avoid making the affected person repeatedly retell the experience to separate departments. Security, people, privacy and communications colleagues should work together, with a named coordinator responsible for keeping the person informed and following through on agreed actions.
Offer support matched to the individual circumstances. Options might include changes to public-facing duties, advice about exposed contact details, help responding to impersonation or access to professional assistance. Discuss any changes rather than imposing measures that feel punitive. Consider family concerns where relevant, and be clear about what the organisation can realistically provide. Do not promise that all copies of information can be removed or that the risk has disappeared.
Review the situation after the immediate publicity fades. Ask whether harassment continues, whether work arrangements remain appropriate and whether the person knows whom to contact if something changes. Communicate with empathy while protecting confidentiality and preserving the investigation. My view is that people should not have to demonstrate visible distress before receiving help. An incident plan focused entirely on systems misses an important part of the job. Nobody should become their own incident manager while everybody else discusses reputational risk.
Under the UK GDPR, a controller must notify the relevant supervisory authority without undue delay and, where feasible, within 72 hours of becoming aware of a personal-data breach, unless it is unlikely to result in a risk to individuals’ rights and freedoms. Where high risk is likely, affected people must also be informed without undue delay, subject to applicable exceptions. The ICO’s guidance explains the thresholds and documentation requirements.
This is not a universal deadline for every cyber alert. First establish whether personal data is affected and assess the consequences for people, including loss of availability or improper alteration as well as disclosure. Involve privacy and legal specialists early. A processor must notify its controller without undue delay after becoming aware of a breach. Establish the organisation’s role rather than assuming every supplier and customer has the same responsibilities.
Keep a record of when awareness arose, what is known, what remains uncertain, the assessment undertaken and the decision-maker. Document a decision not to notify as well as a decision to report. Do not wait for an immaculate final investigation before considering the obligation. Where information is incomplete, seek advice on the appropriate initial notification and subsequent updates. A beautiful report delivered too late is not an improvement on a timely, properly qualified account.
Maintain a notification matrix covering other applicable regulators, contractual counterparties, commissioning bodies and insurers, with triggers and fallback contacts. Different obligations may require different information or timing, so avoid treating the ICO assessment as the whole exercise. Test out-of-hours access to the people authorised to decide. These pages provide general guidance, not case-specific legal advice. The practical objective is a prompt, defensible decision that protects individuals, rather than reporting everything indiscriminately or allowing uncertainty to become an excuse for delay.
Say what you can verify and what people can use. I would explain the confirmed service impact, the action under way, any available alternative and when the next update is due. Distinguish an outage from a confirmed intrusion or data breach. If the evidence is incomplete, say so plainly. “We do not yet know” can be a responsible statement when accompanied by practical assistance and a credible next step.
Prepare a clear route for technical and operational colleagues to validate statements quickly. Identify who approves publication and who acts in their absence. Keep the verified incident picture consistent, but tailor the practical detail: customers need service alternatives, employees need instructions, partners need coordination and journalists need a reliable contact. Consistency does not mean sending every audience the same paragraph and assuming their different questions have somehow been answered.
Choose a publishing route that remains accessible if the affected service fails. Test who can update it and how users will recognise it as authoritative. Include timestamps and correct earlier information visibly when the evidence changes. Avoid repeating slogans, linking to hostile channels or filling gaps with guesses about attribution. Where personal information may be affected, coordinate with the privacy assessment without making legal caution a reason for total silence.
Set an update rhythm that staff can sustain, and honour it even when the update is that investigation continues. Do not promise a restoration time the operational team cannot support. Acknowledge inconvenience and potential harm without making assumptions about everybody’s experience. My preference is for a useful, modest statement over an elaborate declaration of confidence. People need to know how to obtain help and what happens next. A holding statement should hold their confidence, not merely occupy a page while the organisation searches for reassuring adjectives.
Lawful Activism, Monitoring and Ethical Boundaries
Look at conduct, evidence and authorisation, not whether management finds the cause agreeable. Criticism, campaigning and awkward questions are not, by themselves, evidence of a cyber threat. Conversely, sympathy for a cause does not establish that access or interference is authorised. The article distinguishes lawful engagement from unauthorised acts and recommends legal advice on the particular circumstances and jurisdiction. Preserve that distinction in operational decisions as well as public language.
For a specific concern, describe what actually occurred. Was there a threat against a person, an attempt to obtain access, disclosure of information or disruption to a service? Who observed it, and what evidence is available? Avoid categories based simply on political opinions or membership of a campaigning community. A security assessment should explain the relevant behaviour and potential harm, rather than presenting disagreement itself as suspicious.
I oppose colonialism, historic and current, and believe institutions should face scrutiny over the effects of their decisions. That conviction does not require me to excuse exposing ordinary people’s information or interrupting services they depend on. It means assessing both grievances and methods honestly. An organisation can take a complaint seriously while resisting harmful conduct, and it can defend its systems without pretending its own policies are beyond criticism.
In practice, keep stakeholder engagement and security response coordinated but distinct. Give genuine concerns an appropriate route to be heard. Define monitoring purposes narrowly, seek advice where individuals’ rights are affected and maintain clear escalation criteria for credible threats. Review whether labels used in briefings unfairly blur critics with attackers. My judgement is that security is more legitimate when it can explain these boundaries. It should protect people and services, not become a convenient device for avoiding an argument the organisation would rather not have.
Only for a defined, lawful and proportionate purpose. I would begin by identifying the decision the information could change. Are we checking a specific threat, verifying a claim about our services or investigating possible exposure of people’s information? If nobody can explain the operational question, adding another monitoring feed is unlikely to solve it. Curiosity, reputational discomfort and dislike of criticism are not adequate substitutes for a defensible purpose.
Set a clear scope with the relevant legal and privacy advice. Identify suitable sources, who may access collected material, how reliability will be assessed and how long information is needed. Focus on relevant conduct and evidence rather than compiling broad records about people’s political beliefs. Do not equate an active campaigning community with a hostile network merely because its members are persistent or inconvenient. Necessary scrutiny can be uncomfortable without becoming malicious.
Protect the people doing the work. Do not ask an unprepared employee to enter hostile spaces, contact suspected actors or handle suspect material because the organisation wants a quick answer. Establish approved ways of obtaining specialist assistance and preserving relevant evidence. Where a credible threat concerns an individual, coordinate with the appropriate safeguarding, people or security support rather than treating the material as an interesting intelligence item.
Review whether collection produces useful decisions. Has it supported a justified protective measure, clarified a claim or changed the incident assessment? Remove unnecessary collection and avoid retaining unrelated personal information simply because storage is available. Explain the limits of findings in leadership briefings. My recommendation is decision-led intelligence, not surveillance by accumulation. The aim is to understand relevant threats while respecting lawful dissent. A large archive of criticism is not proof of either a sophisticated adversary or a well-run security operation.
Governance, Exercises and Resilience Assurance
The board or trustees should provide strategic oversight, support necessary decisions and challenge the quality of management’s assessment. Start with service consequences and people at risk, not a tour of technical alerts. Ask what is confirmed, what remains uncertain and which decision is needed next. Agree a briefing rhythm that reflects the incident’s seriousness without dragging those doing the work into continuous presentations.
Clarify the boundary between oversight and command. The appointed incident lead should coordinate the response, with delegated technical and operational authority understood. Directors can ask whether options have been properly assessed without selecting network settings or bypassing the response structure. If a decision exceeds management’s authority, make the escalation route explicit. Ambiguity about who may approve spending or accept a prolonged interruption is best resolved before everyone is tired.
Useful board questions concern priorities and trade-offs. Are urgent users receiving help? Is the fallback sustainable? What evidence supports reassurance about information or safety? Which obligations require assessment, and do teams have the resources and specialist advice they need? Consider the capacity of staff handling customers as well as investigators. A technically contained incident may still leave serious operational or human consequences requiring executive attention.
Record major decisions, the available evidence and the reasons for accepting or rejecting options. Avoid judging early decisions solely by facts discovered later. After recovery, require a proportionate account of outcomes, unresolved risks and corrective actions, with owners and follow-up. My preference is for a board that makes difficult priorities clear and insists on evidence, rather than one that demonstrates engagement by joining every operational call. A crowded incident room does not become a better governed incident simply because more senior job titles have arrived.
Test the decisions and dependencies most likely to expose a gap between the plan and reality. As an illustrative scenario, combine a service outage with an unverified data-theft claim, delayed provider assistance and pressure on a public-facing employee. These are exercise ingredients, not a prediction that every incident will include them. Choose details relevant to your services and the consequences you most need to manage.
Give participants incomplete information and require actual decisions. Who declares the incident, approves emergency expenditure and authorises a service alternative? What can be said publicly, and who verifies it? Which evidence would trigger a privacy investigation or a different operational response? Avoid supplying perfect answers through the exercise narrative. The value lies in discovering how people handle uncertainty, disagreement and unavailable decision-makers.
Demonstrate practical arrangements where safe. Contact providers through the agreed exercise process, retrieve the plan without normal systems and walk through a customer’s alternative route. Include accessibility, urgent cases and the staff needed to manage demand. For operational technology, use a separate engineer-led scenario that respects safety constraints and does not disrupt live processes. An exercise should reveal a weakness without creating the incident it was supposed to help prevent.
Finish with a short, candid review. Distinguish a missing capability from an unclear instruction, an untested assumption or a decision outside someone’s authority. Give each corrective action an owner, priority and review date, and retest important failures. Do not turn the debrief into a search for a convenient individual to blame. My aim would be practical improvement, not universal applause. A useful exercise leaves a few uncomfortable but manageable tasks. A certificate confirming that everyone attended is a rather less demanding achievement.
Measure whether important outcomes can be sustained and recovered, rather than assuming that activity equals capability. I would look for exercised fallbacks for priority services, demonstrated provider escalation, controlled administrative access and completed corrective actions that have been retested. These measures should answer a leadership question: what evidence gives us confidence that people can obtain the service when normal arrangements fail?
Define the scope and quality threshold for each measure. If reporting the proportion of services with tested alternatives, identify the service population and what counts as a successful test. A contact-list review is not the same as completing a fallback journey. Specify the test date, relevant limitations and any changes since then. Otherwise, a reassuring percentage may combine very different levels of evidence under the convenient heading “covered”.
Include consequences and unresolved gaps in the discussion. Which high-priority service still has an untested dependency? Which alternative cannot support urgent users or sustained demand? Who owns the risk, and what decision is required? Report important failed tests alongside successes. Their discovery may indicate that assurance is working, provided action follows. Suppressing awkward results makes the dashboard tidier while leaving the organisation no better prepared.
Use attack counts, alert volumes and training attendance as context where useful, not as standalone proof of resilience. They may describe work performed without showing the outcome achieved. Review the measures after exercises, service changes and incidents so that they remain relevant. My preference is for a small set of understandable indicators supported by evidence, rather than a decorative wall of green. Ask a service owner to explain what each measure means for a customer. If that translation is impossible, reconsider whether the measure belongs in the leadership briefing.
Use the article’s 90-day programme as an adaptable delivery sequence, not an official deadline or a promise that resilience can be completed in three months. The intended result is clearer ownership, tested arrangements and honest decisions about remaining risk. Before starting, name an accountable sponsor and agree how progress will be reviewed. Keep the work linked to priority services rather than allowing it to become a separate paperwork campaign.
During days 1–30, identify important outcomes, tolerable disruption, service owners and key dependencies. Verify responsibility for internet-facing assets and provider protection. Agree who handles incident decisions, communications and notification assessments, including deputies. Produce a concise picture of immediate gaps and address obvious exposure through the appropriate technical and operational processes. Do not postpone worthwhile action until every asset has been described in an elaborate inventory.
During days 31–60, demonstrate that the arrangements work. Test provider escalation, independently accessible communications and at least one essential service’s fallback. Run a leadership exercise with incomplete information and record failed assumptions. Where operational technology is involved, use engineer-led work that respects safety constraints. Set priorities for corrective action according to potential harm, not simply how easily an item can be marked complete before the next meeting.
During days 61–90, retest important failures, close priority gaps and present remaining trade-offs to the board or trustees. Record risk acceptance with an owner and review date, and seek decisions on resources that management cannot authorise alone. Explain the evidence behind progress and the limitations still present. The programme should establish an ongoing review habit after the initial period. My test is straightforward: can the organisation demonstrate better continuity and clearer decisions? A newer presentation, however attractive, is not an adequate substitute.
Conclusion: Put the Answers to Work
The useful next step is not to memorise these answers. Choose the questions that expose uncertainty in your own organisation, bring the responsible people together and ask for a demonstration. Start with an important service, its owner, its provider and its fallback. Decide which gap needs action, who will address it and how you will know the improvement works.
Keep the distinction between evidence and confidence visible. Revise an assessment when the facts change, support people whose needs are not captured by a system status and resist the temptation to make every incident a contest with the attacker. Good cybersecurity leadership is practical, curious and accountable. It protects outcomes without claiming omniscience, and it leaves room for a straightforward sentence that is sometimes the beginning of sound judgement: “We should check that.”
Discussion Prompts: Turn Assurance into Action
- Who decides when the usual decision-maker is unavailable? Choose a priority service and imagine it fails outside office hours. Who can declare an incident, approve emergency expenditure and activate the fallback? Name the deputies, identify any authority gaps and agree who will resolve them before the next exercise.
- What would a realistic exercise prove? Take one essential customer task and walk through how it would be completed without the main service or normal sign-in. What evidence would demonstrate that the alternative works, including for people needing additional support? Agree a safe test, its owner and a date, then identify which failed assumptions would require urgent action.
- What evidence would change our view of preparedness? Select one reassuring dashboard measure and ask what sits behind it: which services, which test and which unresolved limitations? Decide what improvement should be demonstrated during the 90-day programme, who owns any remaining risk and when the board or trustees will review the evidence. A green square is welcome, but it should bring its working with it.
