Cybersecurity

Ransomware Negotiation, Explained: What Actually Happens

How ransomware negotiation really works: proof checks, sanctions screening, and why preparation now matters more than bargaining skill later.

Editorial Team / /10 min read
An hourglass with sand running down stands between two olive rotary phones on a table, a tense negotiation against time

When a company discovers its files are encrypted and a ransom note is sitting on every desktop, the instinct is to treat it as a single, terrifying decision: pay, or don’t. In practice, that moment is the start of a structured, checkable sequence, not a one-shot gamble. The ransomware negotiation process a professional runs does not look like haggling at a flea market. They verify who they’re actually dealing with, test whether the claimed access is even real, and screen the whole thing against sanctions law before a cent moves. Understanding that sequence matters less for the rare organization that ends up negotiating, and more for everyone else, because it reveals where the real advantage sits: not in the negotiation room, but in the weeks and months before the attack even starts, much like the zero-trust security principle of assuming a breach before it happens rather than reacting once it has.

The negotiation is a verification process, not a conversation

Belgian cybersecurity firm Secutec, founded in 2005, has built a specialty around this exact work. Its founder, Geert Baudewijns, has been negotiating ransomware cases since around 2016 and says he has worked more than 600 of them, a track record detailed in his book Négocier dans l’ombre (published in English as Negotiating in the Dark, Lannoo, 2025). Speaking on the French-language tech podcast Underscore_, he described negotiation less as persuasion and more as a series of checks a negotiator runs before anyone discusses money.

The first check is whether the group can actually decrypt anything at all. Some extortion attempts are opportunistic: a low-skill actor claims to have encrypted a network and demands payment, but has no real access or decryption capability. A negotiator’s job is to force proof before taking the threat seriously, the same way a buyer would ask a seller to demonstrate a product works before paying for it.

Baudewijns says professional ransomware groups increasingly use what he calls a “file tree” mechanic: before any proof of decryption is offered, the attacker sends a list of the exact filenames they claim to have stolen. If that list matches real files that exist on the victim’s network, it is a strong, though not conclusive, signal that the intrusion is genuine and the group has real access, rather than a bluff. This detail is specific to Baudewijns’s own case history and has not been independently verified elsewhere, but it illustrates the broader point: serious negotiation opens with verification, not bargaining.

Once a group is confirmed as credible, the process moves to a second, entirely separate check that has nothing to do with the attacker’s technical claims. The same logic drives how security teams now sequence frameworks like OWASP, MITRE, and NIST when assessing a different kind of risk: a defined order of checks beats an ad hoc judgment call every time.

Before any ransom discussion goes further, a negotiator or the victim’s legal counsel has to determine whether the group behind the attack is on a government sanctions list. In the United States, that means checking against the Office of Foreign Assets Control (OFAC) list; in the EU and UK, equivalent sanctions regimes apply. If the attacking group, or the wallet address it is demanding payment to, is tied to a sanctioned entity, paying the ransom can itself be a legal violation for the victim, regardless of how badly they want their data back.

This is not a formality tacked onto the process. It is often the step that determines whether payment is even on the table, independent of price, timeline, or how convincing the attacker’s proof looks. A negotiator’s real value here is partly legal risk management: confirming who is actually being paid, not just what they’re demanding.

A locked computer monitor glowing red in a dark room, displaying an encrypted file directory during a ransomware incident

With legality and credibility both checked, what remains is the part people assume is the whole story: agreeing on an amount and a timeline. In Baudewijns’s account, that part has changed dramatically, and not in the direction most people expect.

The negotiation itself has gotten dramatically shorter

Baudewijns says the negotiation phase, the back-and-forth over price once both sides are talking, has compressed from roughly 10 to 15 days a few years ago to about 3 to 4 days today. He attributes this partly to attackers themselves operating on tighter internal timelines and partly to negotiators on both sides having more established playbooks than they did earlier in the ransomware era. That figure comes from his own caseload and has not been independently corroborated, so it should be read as one experienced practitioner’s observation rather than an industry-wide statistic.

This is a separate number from a related but distinct trend, reported by the same negotiator: how long attackers sit inside a network before they even trigger encryption. That figure, often called dwell time, has compressed even more sharply, from weeks or months a few years ago to a few days today, with few cases lasting longer than six days, according to Baudewijns in an interview with ITdaily. These are his figures, not an independent measurement. He points to a market in which initial access is sold on to several ransomware gangs (the so-called Initial Access Brokers), criminals who specialize in breaking into networks and then selling that already-compromised access to ransomware operators, who buy it and move straight to encryption rather than spending weeks exploring the network themselves.

Put together, these two numbers describe the same underlying shift from different angles: the entire ransomware timeline, from initial break-in to final negotiation, has collapsed from weeks or months to days. That compression is the actual durable finding here, more than either individual figure, because it changes where an organization’s response time has to come from. If attackers are inside and ready to encrypt within days of buying access, there is no longer a comfortable window to improve detection after the fact; that window has to already exist before the intrusion happens. It is the same reasoning behind why software supply chain security has shifted from reactive patching toward verifying trust upstream, before a compromised component ever reaches production.

A server-room patch panel with dense cabling and a single illuminated red port, evoking network access changing hands between an Initial Access Broker and a ransomware operator

What ransom demands actually look like

Public reporting on ransomware tends to lead with headline-grabbing figures from the largest attacks on hospitals or critical infrastructure, but Baudewijns describes a much wider and, for most businesses, more mundane range. Drawing on that caseload, he says demands for small and mid-sized businesses typically fall between $30,000 and $100,000, while larger organizations see demands in the $250,000 to $300,000-plus range, with amounts above $500,000 to $600,000 being comparatively rare in his experience. He also says demands are often pegged loosely around 10 to 12 percent of the victim’s annual revenue, suggesting attackers do at least some reconnaissance on a target’s finances before setting a price, rather than picking a number at random.

None of these figures are independently audited statistics; they are one negotiator’s pattern from his own experience. What they usefully illustrate, regardless of the exact numbers, is that ransomware economics function less like a single catastrophic shakedown and more like a pricing model calibrated to what a given victim can plausibly pay, which is itself a reason preparation and cyber-insurance planning matter more than any single negotiating tactic.

Negotiators like Baudewijns are typically engaged by the victim’s cyber-insurance carrier, which he describes as his main clients; the decision to pay stays with the victim. Baudewijns puts his own fee at roughly €3,000 to €4,000 for smaller, more straightforward cases up to €15,000 to €20,000 for larger or more complex ones. That detail matters for a reader evaluating preventive security spending: the cost of professional negotiation is itself a recurring, budgetable line item, not a one-time emergency expense, and organizations that maintain cyber-insurance are effectively pre-purchasing access to this expertise before they ever need it.

What happened in Antwerp, and why the aftermath stayed murky

The clearest real-world illustration of how negotiation intersects with public accountability comes from the city of Antwerp, Belgium. On 6 December 2022, the Play ransomware gang hit the city through Digipolis, the municipal partner that handles the city’s digital and IT services. The attack disrupted municipal services, schools, police, fire department systems, and childcare and elder-care services. The group claimed to have stolen 557 gigabytes of data, which BleepingComputer reported as including passports, IDs and financial documents, and Belgian broadcaster VRT NWS reported a 19 December 2022 deadline.

What happened after that deadline remains genuinely unresolved. Antwerp’s mayor, Bart De Wever, stated publicly that the city never negotiated and never paid a ransom. Some commentators have publicly expressed skepticism about that account, but no independent evidence has surfaced to confirm or contradict it either way. That makes the mayor’s statement a disputed official position, not a settled fact in either direction, and that ambiguity deserves stating plainly rather than resolving one way for the sake of a cleaner story.

A separate rumor also circulated around the case: that an unnamed third party quietly paid a ransom specifically to prevent the leak of documents connected to the Port of Antwerp. Baudewijns raised this claim himself on the podcast, and just as quickly dismissed it as a myth with no evidence behind it. It appears here only because it illustrates how much speculation tends to fill the vacuum when an official account is incomplete, not because there is any reason to treat it as true.

Where AI fits, according to one negotiator

Given how often artificial intelligence is credited with reshaping cybersecurity, Baudewijns’s own view on its role in ransomware specifically stands out: in his view, AI has not meaningfully changed the mechanics of how ransomware attacks are carried out on the attacker’s side, at least not yet. This is presented here as his personal assessment, not as an established industry consensus, and other reporting on ransomware operations may well describe AI playing a larger role in areas like phishing lure generation or reconnaissance. The point of including it is not to settle the question, but to flag that even a practitioner with 600-plus cases behind him is cautious about overstating AI’s current impact on the specific mechanics he deals with daily: encryption, exfiltration proof, and price negotiation.

The FAQ people actually ask about ransomware negotiation

Is it illegal to pay a ransomware gang?

Paying a ransom is not automatically illegal, but it can become illegal if the attacking group or its payment wallet is tied to an entity on a government sanctions list, such as the US Office of Foreign Assets Control (OFAC) list. This is why sanctions screening happens before any payment discussion, not after.

Do negotiators actually get the ransom lowered?

Yes, ransomware negotiators routinely reduce the initial demand, sometimes substantially, but the bigger value they add is verification: confirming the group can actually decrypt the data and that paying would not violate sanctions law, before price becomes the main question at all.

How long does a ransomware negotiation typically take?

According to negotiator Geert Baudewijns, the back-and-forth negotiation phase itself, once both sides are actively talking, has shortened from roughly 10 to 15 days a few years ago to about 3 to 4 days today, reflecting both attacker urgency and more established playbooks on both sides.

Who actually pays for a ransomware negotiator?

A victim’s cyber-insurance carrier typically engages the ransomware negotiator, rather than the victim hiring one alone, which is one reason cyber-insurance coverage is treated as a preventive investment rather than just a payout mechanism.

Can a company always trust a ransomware group to decrypt after payment?

No, and this is exactly why verification precedes payment rather than following it. Professional groups that depend on repeat “business” tend to honor decryption to maintain their reputation among other victims, but there is no legal guarantee, which is why negotiators push for proof before any money moves.

#ransomware#incident-response#cybersecurity#negotiation#cyber-insurance