Six Things Every Exec Believes About Recovery, Until They’re Wrong | #hacking | #cybersecurity | #infosec | #comptia | #pentest | #ransomware


Within the first hour of almost every ransomware recovery Fenix24 runs, someone on the client side says the backups are fine. It’s usually the infrastructure lead, a few hours into the worst night of their career, and they say it to steady the room.
They believe it. So does their CIO. So did the board that approved last year’s resilience budget. Nobody in that chain ever watched those backups bring a working business back.
Accenture’s 2026 study found 81% of executives expect ten days or less of critical downtime from a serious incident, while Accenture puts average actual recovery at three to six months. The UK’s Cyber Security Breaches Survey counted continuity plans at 85% of large businesses and described them as “largely untested or incomplete.” In Fenix24’s engagements, our clients’ documented recovery time objectives usually promise 24 to 48 hours. Across more than 800 clients, four came close for part of the business, and none reached full operations for weeks.
The backup sentence is the first of six assumptions we hear at the start of nearly every engagement. Four of them overestimate readiness. The other two come from caution, and they cost just as many days.
1. “My backups are good.”
“Good” usually means the jobs report success. A green dashboard says a job ran. It says nothing about what happens when an attacker with domain admin rights goes looking for that backup.
In 94% of the environments we recover, the backup system is joined to the same production directory the attacker compromised first, so the credentials protecting the backups belong to the attacker too. Even when backups survive intact, 38% of the time they can’t carry the recovery. The data is too old to resume operations on, was corrupted long before the attack, is the wrong type for the workload, or restores more slowly than rebuilding from scratch. We regularly find backups labeled immutable sitting on infrastructure that cannot enforce immutability.
A backup that exists, a backup that survives, and a backup you can run the business on are three different things. Most organizations have only verified the first.
2. “Everything is backed up.”
Coverage gets measured by server. Recovery happens by business service. The ERP server is in the backup policy, but the database it reads from, the license server it checks in with and the DNS records that let users reach it often aren’t. Restore the server alone and you get a machine that boots and does nothing.
Some components can’t be restored at all. A legacy operating system or an application version the vendor stopped selling becomes a procurement problem the moment the only working copy is encrypted.
The check is easy to describe and rarely done. Pick one revenue-producing service, trace every component it needs to function, and compare that list against the backup policy line by line.
3. “We know our applications and the order they restore in.”
None of the organizations that Fenix24 has worked with arrived at a recovery with a complete map of their applications and dependencies. The closest versions lived in a configuration database that was encrypted along with everything else or got pieced together in the war room while the business was down.
Restore order is set by dependencies, and the first dependency is almost always identity. If the directory can’t be trusted, nothing that authenticates against it comes back cleanly. In 99.2% of the organizations we recover, there is no documented identity recovery plan. The few that existed didn’t survive contact with the attacker. Teams spend the first days rebuilding the directory before a single business application returns. Almost no runbook starts there.
4. “We know what is on our network.”
Attackers work in the parts of a network nobody inventoried, like the hypervisor host stood up for a project three years ago, the out-of-band management cards, or the backup console reachable from the user VLAN.
In 95% of our engagements, critical infrastructure consoles had no meaningful multifactor protection, while only 15% lacked adequate controls at network ingress. Organizations guard the perimeter carefully and leave the management plane (which controls every virtual machine and every backup) behind a single password.
5. “Everything must be patched and hardened before the business can operate again.”
After an incident, the instinct is to bring the entire estate up to a standard it never met before anything reopens. That ties the restart to the slowest item on a remediation list hundreds of lines long, and an outage of days becomes one of months.
Recovery works in sequence. The services the business cannot operate without come back first, into a clean environment, with credentials rotated, the attacker’s access paths closed and monitoring in place. Those systems are hardened before they return. Everything else gets remediated in priority order while the business runs.
6. “The ransomware will keep spreading even after containment.”
This fear keeps systems dark after the threat is dealt with. Most modern ransomware is human-operated. It moves through stolen credentials and legitimate admin tools, and encryption runs because an operator pushes it. Cut off that access, reset the identity infrastructure and remove persistence, and the payload has no path to new systems.
The risk after containment is reinfection through a credential, scheduled task or remote access tool the investigation missed. That is a scoping problem that thorough forensics can answer. Treating contained ransomware like an airborne virus leaves the business offline while the evidence says it could be running.
The test none of these beliefs survives
Each of the six starts as a reasonable assumption. They get expensive because none is tested until the day it matters.
Run a full restore of one critical business service, end to end, against your stated recovery time. Trace its dependencies, starting with identity. Write down what you couldn’t bring back and how long the rest took. That number will be worse than the one in the plan, but it will be far more useful.

Join our LinkedIn group Information Security Community!

——————————————————-
Click Here For The Original Source.