Key Takeaways
- Executives expect to recover from a cyberattack in about 10 days. Real-world recovery takes three to six months on average, according to new research published in Harvard Business Review.
- The organizations that bounce back fastest plan to keep a “minimum viable enterprise” running while their systems are still down or can’t be trusted.
- Five moves get you there: define your operating core, decide who decides, map your dependencies, build a clean place to recover, and test it all under realistic pressure.
- It’s the same recovery-first thinking that shapes how Emerge approaches cybersecurity, backup, and incident response for our clients.
Here’s a question worth putting to your leadership team this week. If every application your company runs went dark at 8 a.m. Monday, what would absolutely need to be working by Thursday?
Most organizations can’t answer that quickly. And according to a Harvard Business Review article published in October, that gap is one of the biggest reasons cyberattacks turn into months-long ordeals.
The authors, a team of global cybersecurity and resilience leaders, surveyed 1,000 senior executives across 20 industries and 13 countries. What they found should make any CFO a little uneasy. Eighty-one percent of executives believe their critical systems would be restored within 10 days of an attack. The authors’ analysis of actual cyber disruptions puts the average closer to three to six months. Meanwhile, 73% still believe their current “predict-and-control” approach to security is sufficient.
Why the Old Continuity Playbook Comes Up Short
Most business continuity plans were built for floods, fires, and power outages. They assume that once the crisis passes, you’ll restore operations from systems you can trust. Ransomware doesn’t play by those rules. Attackers often sit inside a network for weeks before anyone notices them, which means the servers and backups that look fine on the surface may already be compromised. Restore from the wrong copy, and you can reinfect the business you’re trying to save.
The attacks themselves are also getting faster. The 2026 Verizon Data Breach Investigations Report (DBIR) found that exploiting software vulnerabilities has overtaken stolen credentials as the most common way attackers get in, a first in the report’s 19-year history. Ransomware showed up in 48% of breaches, and the median time for organizations to fully patch a vulnerability after detection actually grew, from 32 days to 43, according to Help Net Security’s summary of the report.
And the costs pile up in a predictable place. NetDiligence’s 2026 Cyber Claims Study, released in September, found that among small and midsize enterprises (SMEs), incidents involving business interruption cost, on average, more than five times as much as those that didn’t. SMEs, which the study defines as companies with less than $2 billion in annual revenue, accounted for 97% of all claims.
You don’t have to look far for examples. When a cyberattack hit medical device maker Boston Scientific on August 25, it disrupted order processing and shipping, and employees at the company’s manufacturing facilities in Cork, Ireland, were sent home. Two weeks later, key systems were back and backlogged orders were moving, but the company told investors it couldn’t yet provide a timeline for full operational recovery and expected a material hit to its third-quarter and full-year results. In July, a ransomware attack forced Coca-Cola’s Fairlife dairy unit to suspend production at its U.S. facilities.
Both are large companies with substantial security budgets, and neither could simply flip a switch and get back to normal. For a mid-market manufacturer, medical practice or financial firm with a fraction of those resources, deciding ahead of time how the business will run during the disruption matters even more.
Meet the “Minimum Viable Enterprise”
The HBR authors give that idea a name: the minimum viable enterprise. It’s the smallest set of decisions, applications, and processes your organization needs to keep serving customers, protecting employees, meeting regulatory obligations and bringing in revenue during the first hours and days of a crisis.
Companies that define it in advance don’t waste the first critical day arguing over what to restore first, who has the authority to make the call, or which manual workarounds to switch on. They already know.
The article lays out five moves for getting there. Here’s each one, along with what it looks like in practice for the organizations we work with.
1. Define Your Operating Core
Ask every business unit the same question: if all applications were down for 72 hours, which outcomes would cause the greatest harm if they stopped?
The answer should be framed around outcomes rather than technology. For a manufacturer, that might mean shipping orders that are already built and keeping safety systems on the plant floor running. For a healthcare provider, it’s patient safety. The HBR authors point out that many hospitals that recovered quickly from the 2024 global IT outage had already identified which clinical applications were essential and what manual protocols to activate when digital tools failed. Those without fallbacks for imaging, patient records, and emergency communications faced far greater clinical risk. For a financial services firm, it could be client access to accounts and on-time regulatory reporting.
This goes a step beyond a traditional business impact analysis (BIA). A BIA tells you what downtime would cost. The 72-hour exercise tells you what the company still has to be able to do once the disruption is underway.
2. Decide Who Decides, Before You Need To
In a cyberattack, the limiting factor is often decision-making speed rather than technical skill. Who can take a customer-facing system offline? Who decides when to switch to paper? When do you notify customers, regulators, suppliers, or your cyber insurance carrier, and who makes that call?
Only 43% of executives in the HBR survey said cyber resilience is governed as a business outcome shared across leadership. That leaves a lot of organizations treating recovery as an IT problem right up until it becomes everyone’s problem.
“The technical side of a ransomware event, we can work through that. What slows everybody down is those first few hours when nobody’s sure who gets to make the call. Do we shut down the enterprise resource planning (ERP) system? Do we call the insurance carrier first or the attorney? Who tells the customers, and what do we tell them? I’ve sat in those rooms, and if the answers aren’t written down somewhere ahead of time, you can lose half a day just figuring out who’s in charge,” said Jeff Aiken, Emerge’s Cyber Security Manager.
3. Map the Dependencies that Could Stop You Cold
Sixty-nine percent of organizations in the HBR study said their operations increasingly depend on complex internal and third-party digital ecosystems. Only 38% have a clear view of their most critical interconnections.
That blind spot is getting more expensive. The 2026 DBIR found that third-party involvement in breaches jumped 60% year over year and now accounts for nearly half of all breaches.
Dependency mapping means looking past the obvious applications. Which identity systems do your critical functions rely on? If Microsoft 365 is unavailable, how do your people reach each other? Which cloud services, data feeds, and suppliers sit underneath your operating core? A quick tabletop walk-through usually turns up at least one surprise, often something as simple as a shared login or a vendor everyone forgot about.
4. Build a Clean Place to Recover
This is where cyber recovery differs most from traditional continuity planning. After a ransomware attack, your primary environment may not be trustworthy. You need a known-good foundation to run your essential core from, built on validated, immutable backups that are stored separately from the systems they protect and tested under real conditions.
This is a big part of why our DataWATCH backup and disaster recovery service is built around immutable backups, meaning data that can’t be altered or deleted for a set period once it’s written, and why recovery testing is built into the service from the start. Plenty of companies have discovered during an incident that their backups were incomplete, corrupted, or reachable by the attacker. Finding that out during a scheduled test is a lot less painful.
5. Test Under Realistic Conditions. Then Test Again.
Only 35% of organizations in the HBR survey regularly run real recovery and continuity exercises rather than relying on audits and documentation.
A separate Resilience Benchmark Report from CYGNVS and Marsh, released this month and based on more than 9,500 survey responses, paints a similar picture. Nearly all organizations (92%) have documented incident response processes, but 84% review those plans once a year or less. Seventeen percent don’t run tabletop exercises at all, and only 37% have playbooks written specifically for ransomware.
How This Lines Up With the Way Emerge Works
Reading the HBR piece, we found ourselves nodding along. It describes a lot of what we already believe about cybersecurity for mid-market organizations.
“When we sit down with a client to talk about recovery, I usually don’t start with backups or firewalls,” Aiken said. “I’ll ask the owner or the CFO, ‘Okay, what happens if you can’t ship for three days? What happens if you can’t bill?” Some people have an answer right away, and a lot of them don’t, and that’s fine; that’s why we’re having the conversation. Once we know what the business can’t live without, the technology decisions get a whole lot easier.”
Start With the 72-Hour Question
You don’t need a six-month project to begin. Get your leadership team in a room, ask what would need to keep running if every application went down for 72 hours, and write down the answers. Then ask who would make the key calls. The gaps will show up quickly.
If you’d like a structured place to start, take Emerge’s Data Resiliency Assessment or schedule a conversation with our team about where your recovery plan stands today.
