Skip to main content

Cyber insurance pays for forensics, legal counsel, ransom negotiation and business interruption. It does not pay back the confidence a client loses while your e-mail is down and nobody can tell them what happened. The Centre for Cybersecurity Belgium logged 588 incident reports in the first nine months of 2026, and the gap between a covered loss and a damaged reputation is where most Belgian companies are still unprepared.

What cyber insurance covers and what it leaves standing

Cyber insurance is a policy that reimburses the measurable cost of a digital incident, typically forensic investigation, legal advice, ransom negotiation, data restoration, notification duties and business interruption losses, within an agreed limit. It does not restore the trust a client, employee or regulator loses in the days after the attack. European risk managers flagged exactly this boundary in FERMA’s 2025 review of the cyber market, which notes that cover for reputational harm stays limited and is usually written as a sub-limit for public relations costs rather than compensation for lost future business. The practical reading for a Benelux company is simple. A policy pays for the people who repair the systems and defend the file, and it may fund an external communications adviser, but the words that reach clients and staff still come from the company itself. Whoever writes those words decides what the incident costs in reputation.

A cyber policy usually pays forThe company still owns
Forensics, restoration and legal counselWhat clients are told, and when
Ransom negotiation and sanctions screeningWhat staff hear before reading it elsewhere
A capped crisis communication budgetLost future business, and the spokesperson

Why 588 reported incidents reset the Belgian baseline

The Centre for Cybersecurity Belgium registered 588 cyber incident reports between 1 January and 28 September 2026, already more than the 556 recorded across the whole of 2025, with 220 of them falling in the third quarter alone. System compromise was the fastest-growing category, tripling against the previous quarter, and public administration and healthcare took the heaviest hit. Those numbers describe notified incidents, not attempted ones, so they mark a floor rather than a ceiling. For a Benelux B2B company the operational consequence is a question of odds. An incident that once felt like an outlier now sits inside a normal planning horizon, which means the communication response needs the same standing status as the backup routine. A company that treats disclosure as an improvised task will improvise it in front of clients, staff and a regulator at the same time.

Stef Vermeulen of cyber insurer Stoïk describes the targeting logic plainly. Attackers scan the internet automatically for weaknesses and never read the company name or the turnover first, so the butcher and the baker appear in the same results as a multinational. Stoïk screens IP addresses, subdomains, outdated software, unsecured VPNs, forgotten WordPress installations and leaked credentials before issuing a policy, and just over 60% of applicants pass first time. Underwriting has become a readiness assessment. Communication has not caught up.

The crisis team that has legal and finance but no communications seat

The crisis team an insurer recommends after handling cyber claims names IT, management, HR, legal and finance, and that list leaves out the function that decides how the incident is remembered. Communications is not a department to inform once the technical picture settles. It is the seat that drafts what clients read, what staff hear on day one and what a journalist quotes in week two. IBM’s 2026 Cost of a Data Breach report puts the global average breach at 4.99 million dollars and identifies lost business as the single largest cost component, which is the part no forensic team can repair. It also measures an average of 247 days to identify and contain a breach, a window in which silence becomes its own message. Adding a communications lead to the team costs nothing at the planning stage and changes what the organisation can do in the first hour.

Claims handlers reach the same conclusion from the other end. Lara Vansweevelt of AON Belgium’s claims team observes that the companies which manage an incident best are “not those with the best technology but those with the best internal organisation” and the clearest crisis structure. Technology decides how long the outage lasts. Structure decides what the outage means.

In the crisis plans we build for Benelux clients, communications is named before any technical scenario is discussed, because whoever holds that seat needs standing authority rather than an invitation. That is the principle behind a continuity plan without a communication plan being half a plan. A restored server with no explanation attached is a technical success and a commercial loss.

Who tells clients, staff and regulators what when e-mail is down

The first operational problem in a cyber incident is not the message, it is the channel. When e-mail, the website and the CRM sit behind the same compromised infrastructure, every standard route to a stakeholder closes at once. Every statement also clears the insurer and legal counsel first, because wording can affect both cover and liability. Nobody comments on a ransom demand.

  1. Call the insurer’s emergency line first. Acting alone invites a later dispute about cover.
  2. Activate the crisis team with communications included, not IT alone, and confirm who decides that day.
  3. Publish a holding statement on a channel the attacker does not control, such as LinkedIn, a pre-registered secondary domain, or phone calls.
  4. Brief staff before clients learn anything, because employees are the channel every client already trusts.
  5. Align each regulatory notification with legal counsel, including NIS2 duties where they apply.
  6. Log every message with timestamp, audience and approver, from day one.

Documenting from day one is the insurers’ third lesson from claims handling, and most companies apply it only to systems. The claim file benefits from a message log too, and so does the company six months later when a client asks what it was told and when. The first hours of a crisis are almost entirely a documentation discipline.

Does executive visibility on LinkedIn feed AI voice phishing

It does, and that is still not a reason to go quiet. Large-scale AI phone phishing uses bots that call in a voice and context assembled from public professional profiles, so the LinkedIn presence that builds an executive’s authority also supplies the attacker’s script. Stoïk screens incoming mail and calls against 130 factors because the tells are no longer audible.

The scale is measurable. Cloned-voice vishing rose 1,633% in a single quarter according to Keepnet’s 2026 vishing research, and roughly 40% of business e-mail compromise attempts now involve AI-generated voice, video or text, up from under 5% in 2023. A familiar voice has stopped functioning as proof of identity.

Pulling an executive out of public view would trade a commercial asset for a marginal security gain. The workable response is a verification rule that applies regardless of who appears to be calling. Payment instructions, credential resets and urgent approvals get confirmed on a second, pre-agreed channel every time, and staff are told that insisting on it is correct behaviour. CEO positioning and voice-phishing resilience are compatible, but only if that protocol exists before the first convincing call.

Why cyber risk belongs in the deal communication plan

Whoever buys a company also buys its cyber risk, including attackers who may have been inside the network for months before the term sheet existed. Marco Salvatore of AON Belgium puts the exposure at almost one in four companies facing a cyber incident during or shortly after a merger or acquisition. Independent dealmaker research points the same way, with 53% reporting that cybersecurity issues have put a transaction in jeopardy and technology due diligence rated the most burdensome review element by 51% of respondents in 2026.

An incident landing mid-integration hits an organisation at its least coordinated moment, when two sets of internal channels, two spokespeople and two sets of client expectations are still being merged. Extend the deal communication plan to an incident scenario. Name one spokesperson across both entities, agree who informs which client portfolio, and decide how investors and financial media hear about it. That is the cost described in internal communications during M&A and the cost of silence, compressed into days.

A cybersecurity score is a board document before a technical one

AON’s IQ platform runs a company through more than 100 questions to produce a cybersecurity score between 0 and 4, with a sector benchmark and a scenario-based exposure analysis attached. It is built to be shared with the board, shareholders and auditors, which makes it a communication artefact as much as a security one. A 2.4 means nothing to a non-executive director unless someone explains the benchmark and the two weakest inputs.

Cyber insurance has grown cheaper and broader since the ransomware wave of 2020 and 2021, and Salvatore notes that an SME with 5 million euro turnover often pays under 5,000 euro a year for 1 million euro of cover. One honest limitation belongs in that board conversation. Nobody can yet compare the return on one security measure against another with hard data, so part of the business case still rests on judgement.

Questions Belgian companies ask about cyber insurance and reputation

Does cyber insurance cover reputational damage

Rarely in the way companies expect. Most policies fund crisis communication services up to a sub-limit, covering advisers and notifications. Pure reputational loss, meaning lost future business and brand devaluation, is normally excluded unless it converts into documented losses inside the policy period. The adviser is covered. The lost client is not.

Does a policy still pay out after an employee clicks a phishing link

Generally yes, and assuming otherwise is one of the most common misreadings in the market. Human error, including a clicked link or a misconfigured server, sits within normal cover, and supply-chain attacks reaching you through a supplier are covered by most modern policies. The exclusions that matter concern known unpatched vulnerabilities.

Who should speak publicly during a cyber incident

One named spokesperson, with a named deputy, agreed before any incident. Technical detail travels from IT through that person rather than straight out of IT, so the wording stays aligned with the insurer and legal counsel. A rotating cast of speakers reads externally as an organisation that has lost control of its facts.

Is a Belgian SME really a target

Yes. CCB figures reported by ITdaily in 2026 put Belgium at an average of 22 ransomware attacks a day, with 42% of Belgian SMEs already hit and 30% of companies running without any cyber strategy. Automated scanning does not read turnover figures.

What should a company say in the first hours when nothing is confirmed

Say what is known, what is being done, and when the next update follows. A holding statement needs neither a confirmed cause nor a confirmed scope, and waiting for both guarantees someone else frames the story first. It must never speculate about the attacker, the data, or any ransom demand.

Putting a communications seat next to the insurer’s hotline

A communications seat in the cyber crisis team is the cheapest piece of cyber resilience a Benelux company can buy, because it is a decision rather than a purchase. The insurer’s emergency line handles forensics, legal exposure and ransom negotiation, and the CCB counted 588 notified incidents in the first nine months of 2026, so the call itself is increasingly routine. What stays unresolved after that call is what clients, staff, partners and regulators are told, in what order, on which channel, and by whom. Companies that come out of an incident with their relationships intact are rarely the ones with the best detection stack. They are the ones who had pre-approved holding statements, an offline contact list, a named spokesperson and a message log before anything happened. Build those seven items this quarter and the next incident becomes an operational problem instead of a reputational one.

  • Give communications a permanent seat in the cyber crisis team, alongside IT, management, HR, legal and finance.
  • Pre-approve holding statements for the three most likely scenarios, cleared once by legal counsel and the insurer.
  • Keep an offline contact list for staff, key clients, suppliers and the emergency line, printed and stored outside the network.
  • Name a spokesperson and a deputy, and media train both before anything happens.
  • Log every message from the first hour, with timestamp, audience and approver.
  • Write a verification protocol for payment and credential requests that holds even when the voice is familiar.
  • Rehearse the whole sequence once a year with e-mail and the website assumed down.

None of these seven items needs a budget line. They need one afternoon and a named owner. Companies that rehearse find the same two gaps every time, an out-of-date contact list and no agreement on who may speak. That is a cheap lesson in a meeting room and an expensive one during an incident, and crisis management in the Benelux comes down to whether the structure existed before it was needed.