Skip to content
Security & Trust

Rubrik: forensics takes days, but its recovery cases took a median 20 days

Rubrik Zero Labs data suggests recovery time, not recoverability, is the number a board should ask for

W
WebPulse Newsroom
AI-assisted · 4 min read
Share on X LinkedIn
Rubrik: forensics takes days, but its recovery cases took a median 20 days

Graphic: WebPulse, drawn from the sourced facts in this story.

In brief
  • Rubrik Zero Labs says forensic answers came fast in its cases, but the median organisation it helped needed 20 days to recover, and 28% needed over six weeks.
  • Rubrik recorded attackers trying to sign in to the backup management console in at least 7% of its recovery cases.
  • Leaders should ask for a timed recovery rehearsal and treat backup tools as privileged systems.

Being told what happened is not being back in business

After a ransomware attack, the first answers arrive quickly. The first working day is another matter. Rubrik Zero Labs has published its 2026 ransomware report, and its recovery data shows a long gap between understanding an attack and getting past it.

The thesis of this article is simple. Recovery time, not the existence of a backup, is the real measure of resilience. And the console that controls recovery is a target in its own right. The first point is the publication's argument. The second rests on Rubrik's data.

What Rubrik measured

In Rubrik's cases, forensic answers came quickly. It reports that 45% of findings appeared inside a day and 68% inside two days. Recovery moved at a different pace.

Counting from the day recovery work began, the middle case reached recovery on day 20. Only 32% finished within two weeks. In 28%, it took longer than six weeks.

Rubrik also kept a longer clock, which runs until the customer agreed the case could close. Its middle value was 26 days. In 45% of cases that clock passed 30 days.

Whose numbers these are

Rubrik drew on three views. One is a public tally of 16,953 incidents between January and October 2026, attributed to 100 groups the team checked individually. Another is detections in backup data held by its customers. The third is its own recovery engagements.

The timing figures come from that third view: organisations that asked Rubrik for help. They do not describe every ransomware victim. Rubrik reports medians, so half of those cases ran longer than the middle value.

Why restoration is slow: what the report says and what we infer

The report does not break down where the 20 days go. It tells leaders to rehearse recovery across identity, applications and data. It also says backup scanning helps confirm a clean recovery point.

What follows is our inference, not a Rubrik finding. Those three layers depend on one another. Applications need working sign-in before anyone can use them. Data must be checked before it is trusted. A restored copy that still carries the attacker's tools would invite a second incident.

If that reading is right, the slow part is not copying files. It is deciding, step by step, what is safe to bring back. A paper plan rarely captures that. A timed exercise would.

The recovery console is part of the attack surface

Rubrik recorded attackers trying to sign in to the backup management console in at least 7% of its recovery cases. The field was blank in some cases, so Rubrik says the true share could reach 14%.

The report's own conclusion is that such access deserves the protection given to any privileged system. Our inference follows from that. A console that can restore everything can also be used to damage the ability to restore. Few organisations audit it like a payment system.

Rubrik adds context from its backup scanning. It detected 21 of the 100 active groups in customer backups. Those 21 account for 62% of publicly reported attacks. Rubrik cautions that a missing detection does not prove nothing was stolen.

Theft is common and the way in is often unknown

Rubrik narrowed its public tally to 2,931 reports that say how the victim was pressured. Stolen data appears in 67% of them. Encryption is stated in 45%, and 13% cite both. Rubrik treats these as floors, since authors record only what they chose to describe.

Entry points are murkier. Where a report names one, exploited vulnerabilities lead at 54%, ahead of stolen credentials at 18% and phishing at 14%. Yet 94% of public reports name no entry point. Rubrik concludes that plans should not depend on knowing the cause or the attacker.

What leaders should ask

First, ask for a measured recovery time. Have the team run a full exercise covering identity, applications and data, and record the hours it takes.

Second, ask who can sign in to the backup console, how those accounts are protected, and whether anyone watches them. Rubrik also advises storing recovery points so they are isolated, unalterable and checked for cleanliness.

Third, ask whether the plan holds up if the attacker is never identified. Rubrik points to tested recovery, least privilege and monitored recovery systems as controls that do not need attribution.

Fourth, ask about stolen data separately. A restored system does not undo a theft.

Forensics can explain an attack within days. Only a timed rehearsal shows how long coming back will take.

Produced by the WebPulse Newsroom with AI assistance from the original reporting credited below, and checked against that source by our editorial review. How we use AI.
Original reporting: Rubrik Zero Labs.