At 5:40 a.m., the first-shift supervisor could not open the production schedule. By 6:10, workstations on the shop floor displayed a ransom note, barcode scanners would not authenticate, and the shipping team could not print labels. This manufacturing ransomware recovery example shows what recovery looks like when an attack affects both office systems and the technology that keeps a plant moving.
The scenario is a composite based on common issues seen in small and mid-sized manufacturing environments. The details will vary by plant, equipment, and insurance requirements, but the operational decisions are realistic. The lesson is not that every business can avoid downtime. It is that a prepared business can contain the problem, keep people safe, and restore the right systems in the right order.
The incident: a disruption, not just an IT problem
The company was a 140-person metal fabrication business with one primary facility and a small sales office. Its network supported ERP, file shares, email, engineering drawings, quality documents, shipping, accounting, and several shop-floor workstations. Some CNC machines were network-connected for program transfer, although the machines themselves could continue operating from locally loaded programs.
The attack began with a compromised Microsoft 365 account. The attacker used the account to send convincing messages internally, captured additional credentials, and gained access to a server through an exposed remote-management path. Encryption began overnight, targeting virtual servers, shared folders, and several employee computers.
By the time the first shift arrived, the business faced immediate operational questions:
- Which systems were safe to use?
- Could scheduled work continue without ERP and network drawings?
- Were machine controllers affected?
- Had employee, customer, or defense-related data been copied before encryption?
- Could the company meet same-day shipment commitments?
Those are business continuity questions. They require leadership, operations, IT, legal counsel, insurance contacts, and sometimes law enforcement to work from the same facts.
First four hours: contain the attack and protect the plant
The recovery team did not start by restoring servers. First, it stopped the spread.
IT disconnected affected servers and workstations from the network, disabled compromised accounts, revoked active sign-in sessions, and blocked remote access until it could be reviewed. The team preserved firewall logs, endpoint alerts, ransom notes, and suspicious email evidence. That evidence mattered later for the cyber insurance carrier, forensic review, and determining whether data had left the environment.
Operations isolated the production network from the office network. This was a critical decision. If there is any uncertainty about whether a ransomware event has reached programmable controllers, machine interfaces, or industrial computers, treat the environment carefully. An uncontrolled restart can create safety concerns, corrupt machine programs, or spread malware further.
Meanwhile, managers moved to a manual production board. Supervisors documented active jobs, machine status, material on hand, and work already in process. Shipping created a handwritten list of orders due that day. Customer service used approved personal phone procedures and a clean, separate communication channel to contact customers with time-sensitive deliveries.
Manual work is slower and less accurate than normal systems. But a temporary manual process is often better than waiting for every application to return. The goal was not to make the company fully productive by noon. It was to protect people and preserve the ability to make informed production decisions.
What the company did not do
The company did not reconnect encrypted devices to see whether they would “come back.” It did not let employees use unknown USB drives or personal email to move files. It did not assume backups were clean simply because backup jobs showed as successful.
It also did not negotiate or pay immediately. That decision belongs with executive leadership, legal counsel, the insurance carrier, and incident-response specialists. Payment is never a guarantee of decryption, data deletion, or future safety. In some cases, legal and regulatory restrictions may also apply.
Day one: determine what can be restored safely
Once containment was underway, the recovery team established a clean environment for investigation and restoration. This step separates recovery from wishful thinking.
They reviewed backup records and tested recovery points before bringing core systems online. The team needed answers to three specific questions: Was the backup stored separately from the production network? Could it be restored? Was it created before the attacker gained access?
The manufacturer had immutable cloud backup copies for its virtual machines and a separate backup of Microsoft 365 data. Those copies were not reachable through the same credentials used to manage the primary network. That separation prevented the attacker from deleting every usable recovery point.
A clean virtual server was built first for identity services, followed by network services needed to authenticate users safely. Next came ERP, file services, and the print and shipping systems. Each restored server was patched, scanned, and reviewed before access expanded. Restoring everything at once may sound faster, but it increases the chance of reintroducing malicious tools, compromised accounts, or an unpatched weakness.
The recovery order reflected the plant's actual dependencies. ERP came before broad file-share access because production scheduling, inventory, and shipping needed reliable transaction data. Engineering drawings were restored next, with version checks for active jobs. Accounting could wait a little longer. This is why a recovery plan should be based on business impact, not simply a list of servers.
A manufacturing ransomware recovery example: restoring production in phases
By the end of the first day, the business had clean communications, a controlled production network, and enough manually verified information to run selected work orders. Full ERP access was not yet available, but supervisors could keep priority jobs moving.
On day two, the team restored ERP in an isolated segment and reconciled transactions that occurred after the last clean backup. Inventory quantities, open orders, and shipping records required manual validation. This reconciliation took time, but skipping it would have created a different problem: shipping the wrong material, missing a customer deadline, or making production decisions from incomplete data.
The plant did not reconnect every workstation at once. IT rebuilt and reissued devices in groups, beginning with scheduling, shipping, quality, engineering, and finance. Users reset passwords, enrolled in multifactor authentication, and received direct instructions on how to report unexpected prompts or emails.
By day three, shipping labels and ERP transactions were functioning. Most office staff were working normally, and production scheduling had returned to the system of record. A few nonessential applications remained offline pending review. That was a reasonable trade-off. Restoring a lower-priority tool before it was verified would not improve revenue or customer service enough to justify the added risk.
The company did experience real losses: overtime, delayed shipments, staff time, forensic costs, and customer conversations it would rather not have had. But it avoided a prolonged plant shutdown because recovery priorities had been tied to operations, not convenience.
What made recovery possible
The company was not protected by one product. Recovery worked because several practical controls supported one another.
It had backups that were separate from the production environment, tested restore procedures, documented system dependencies, and a current list of emergency contacts. It also had network segmentation that allowed the office environment and production systems to be isolated without turning every machine into a question mark.
Just as important, leadership knew who could authorize decisions. During ransomware response, delays often come from uncertainty: Who can shut down remote access? Who can approve emergency purchases? Who speaks to customers? Who coordinates with the insurance carrier? Written roles reduce confusion when minutes matter.
For manufacturers with internal IT, the right partner can add monitoring, backup oversight, and incident-response structure without taking control away from the internal team. For businesses without that capacity, managed IT support can provide a defined point of contact before an emergency occurs.
The recovery plan should be tested before an attack
A plan on a shared drive is not proof that a plant can recover. The useful test is a controlled exercise: restore a copy of ERP, bring back an engineering file share, verify a shipping workflow, and confirm that leaders can communicate if email is unavailable.
Testing also exposes uncomfortable but valuable details. A backup may restore successfully but take longer than the business can tolerate. A key application may depend on a server no one documented. A production workstation may use an unsupported operating system that cannot be patched without affecting equipment. These are manageable problems when found on a planned schedule, not during a ransom event.
For Utah and Tennessee manufacturers, Gravity Networks can help turn those findings into a practical recovery plan with monitored systems, documented responsibilities, and regular reviews. The useful standard is simple: know what must keep running, know how long it can be down, and prove that recovery works before the next shift depends on it.
