Ransomware is the cyberattack most people have heard of, because when it hits, everyone notices. Screens fill with ransom notes, hospitals divert ambulances, factories stop and payroll doesn’t run. It’s also more common than many people realize. Verizon’s 2025 Data Breach Investigations Report found ransomware in 44% of the breaches it reviewed, up from 32% the year before, and in 88% of breaches at small and medium-sized businesses.
Ransomware is malicious software that locks or steals an organization’s data and then demands payment to restore access or to keep the data from being published. Modern attacks often combine encryption with data theft, a tactic known as double extortion, and some skip encryption entirely and rely on the threat of leaking stolen data.
In one sentence: Ransomware is digital extortion: attackers get inside, lock up or copy your data, and demand money to give it back or keep it quiet.
Key takeaways
- Ransomware was present in 44% of breaches in Verizon’s 2025 report, and in 88% of breaches at small and medium-sized businesses.
- Attackers usually get in through stolen credentials, unpatched internet-facing systems or phishing.
- Double extortion means encrypting data and threatening to leak it. Some groups now only steal and threaten.
- CISA’s top preparation step is offline, encrypted backups that you regularly test.
- The FBI does not support paying ransoms and asks victims to report attacks either way.
How does a ransomware attack work?
A ransomware attack is rarely a single click that locks everything instantly. It usually unfolds over days or weeks in stages:
| Stage | What the attacker does |
|---|---|
| 1. Initial access | Gets in through a stolen password, an unpatched VPN or firewall, a phishing email or a compromised supplier |
| 2. Foothold | Installs tools to stay connected and avoid detection |
| 3. Privilege escalation | Steals or abuses administrator credentials to gain wider control |
| 4. Discovery and spread | Maps the network, finds file servers, databases and backups, and moves across systems |
| 5. Data theft | Copies sensitive data out of the network to use as leverage |
| 6. Sabotage backups | Deletes or encrypts any backups it can reach, so recovery without paying is harder |
| 7. Encryption and demand | Encrypts systems, often at night or on a weekend, and leaves a ransom note |
This timeline matters for defenders. Every stage before the last one is a chance to spot and stop the attack. Most of the damage happens at the end, but most of the opportunities to stop it come earlier.
How do attackers get in?
The 2025 DBIR found that across all breaches, the leading ways attackers first got in were credential abuse (22%), exploited vulnerabilities (20%) and phishing (16%). Ransomware groups use all three:
- Stolen credentials. Passwords taken by information-stealing malware, reused from other breaches or bought from “access brokers” who sell entry into company networks. Remote access without strong MFA is a favorite target.
- Unpatched internet-facing systems. VPNs, firewalls and other edge devices with known vulnerabilities. Verizon noted that exploitation of these systems grew again in 2025.
- Phishing and social engineering. Emails that deliver malware or steal logins, and phone calls to help desks asking for password or MFA resets.
- Third parties. CISA’s guide warns that managed service providers have been an infection route that affected many client organizations at once.
The good news is that the same basics block a large share of these. Our guide to identity and access management covers phishing-resistant MFA and least privilege, which make stolen credentials far less useful.
What is double extortion?
Early ransomware only encrypted files. Companies with good backups could restore and ignore the demand. Attackers adapted. Around 2019, groups started copying data before encrypting it, then threatening to publish it if the victim didn’t pay. That’s double extortion: pay to get your systems back, and pay again (or as part of the same demand) to keep your data off a leak site.
The CISA, FBI and NSA guide also notes that some actors now skip encryption and simply steal data, using the threat of release as their only leverage. That’s why Verizon counts ransomware “with or without encryption.” Backups alone don’t solve a data theft problem, so prevention and detection matter as much as recovery.
What is ransomware-as-a-service?
Many ransomware attacks are run like a franchise. A core group develops the malware, runs the payment and leak sites and handles negotiations. Affiliates do the actual break-ins and share the ransom with the core group. This ransomware-as-a-service model means attackers don’t need to write their own malware, which lowers the bar and increases the number of attacks.
It also explains why different attacks using the same ransomware can look very different: the affiliate’s methods for getting in vary, even when the final payload is the same.
Why are small businesses hit so hard?
Verizon’s numbers show a sharp gap. In larger organizations, ransomware was part of 39% of breaches. In small and medium-sized businesses it was part of 88%. Many attacks are opportunistic: affiliates scan for any exposed VPN or buy any working password, and whoever is easiest gets hit.
Smaller companies often have fewer of the defenses that slow attackers down. There may be no dedicated security staff, remote access may rely on passwords alone, backups may sit on the same network with the same admin credentials, and nobody may be watching alerts at night. Usually the cause is limited time and budget. The upside is that the most effective controls are also some of the cheapest.
A short list for a small business that wants to start this month:
- Turn on MFA for email, remote access and every admin account, using passkeys or security keys where your services support them.
- Patch your VPN, firewall and other internet-facing devices as soon as fixes are released, and remove ones you no longer use.
- Keep at least one backup copy offline or immutable, with separate credentials, and test a restore every quarter.
- Use separate admin accounts and remove admin rights from everyday user accounts.
- Write a one-page incident plan with phone numbers, print it and keep it somewhere other than the network.
If you use an IT provider, ask them directly how they handle each item, and how their own access to your systems is protected.
How do you prevent ransomware?
The #StopRansomware Guide from CISA, the FBI, NSA and MS-ISAC is the best free starting point. Its recommendations, grouped by purpose:
| Goal | Key controls |
|---|---|
| Be able to recover | Offline, encrypted backups of critical data; test restores regularly; keep a copy of the incident response plan offline and on paper |
| Stop initial access | Patch internet-facing systems quickly; scan for vulnerabilities; require phishing-resistant MFA for remote access and email; train staff to spot phishing |
| Limit spread | Least privilege; separate admin accounts; segment the network; restrict remote desktop and admin tools |
| Detect early | Endpoint detection and response; centralized logging; alerts on unusual admin activity and large data transfers out |
| Manage third parties | Review supplier and managed service provider access; require MFA and logging for their connections |
Backups deserve special attention. CISA stresses that backups must be offline because ransomware actors look for accessible backups and delete or encrypt them. The guide also warns that automated cloud sync isn’t enough on its own, because encrypted local files can sync to the cloud and overwrite good copies. Immutable or offline copies, separate credentials for backup systems and regular restore tests turn backups from a hope into a plan.
What should you do during an attack?
The #StopRansomware Guide includes a response checklist. In short:
- Isolate affected systems. Disconnect them from the network to stop the spread. If you can’t disconnect, power them down, though that may destroy evidence in memory.
- Triage and prioritize. Identify which systems are critical for restoring the business and which are affected.
- Preserve evidence. Capture logs and system images where possible, for investigation and possible law enforcement use.
- Reset credentials. Assume attackers have administrator passwords. Reset them, starting with accounts that control identity and backups.
- Contact the right people. Bring in your incident response team or provider, legal counsel and insurers, and report to law enforcement.
- Restore from clean backups. Rebuild systems and confirm the attacker’s access is removed before reconnecting them.
Practice this before you need it. A tabletop exercise where leaders walk through a ransomware scenario often reveals gaps, such as nobody knowing who can authorize shutting down the network, or the response plan living on the server that just got encrypted.
Should you pay the ransom?
This is a business decision that often lands on executives in the worst possible week. The facts to weigh:
- The FBI does not support paying a ransom. Its reasons: paying doesn’t guarantee you get your data back, some victims never receive working decryption keys, and payments encourage more attacks.
- The FBI also acknowledges that when a business can’t function, leaders will look at all their options.
- Paying doesn’t make stolen data disappear. You’re relying on criminals to keep a promise not to leak it.
- Verizon’s 2025 report found that 64% of victim organizations didn’t pay, up from 50% two years earlier, and the median ransom paid fell to $115,000.
Whatever you decide, the FBI urges victims to report the incident to law enforcement, through a local FBI field office or the Internet Crime Complaint Center (IC3). Reporting can help investigators and sometimes leads to recovered funds or decryption help. Talk to legal counsel early, because reporting obligations and payment rules vary by industry and country.
A worked example: the weekend that didn’t happen
Picture a regional distributor with 300 employees. An attacker buys a working VPN password that came from malware on an employee’s home computer. The VPN has no MFA, so the attacker logs in on a Tuesday night.
Over the next week, the attacker finds an old service account with domain administrator rights and a password that never changes. They use it to reach file servers and the ERP database, copy several hundred gigabytes of customer and HR data to a cloud storage site, and find the backup server, which uses the same domain credentials. On Saturday night they delete the backups and encrypt everything else.
Now replay it with the basics in place. The VPN requires phishing-resistant MFA, so the bought password doesn’t work. Suppose the attacker gets in some other way anyway. The old service account was found in an access review and restricted. Large outbound transfers trigger an alert on Thursday, and the security team investigates. The backup system uses separate credentials and keeps an immutable copy offline. Even if encryption happens, the company restores from clean backups in days instead of weeks, and the early alert gave them a chance to stop the data theft.
No single control would have saved the first company. Several ordinary ones, working together, gave the second company many chances to catch the attack before Saturday night.
Frequently asked questions
What is the most common way ransomware gets in?
Through stolen or weak credentials, unpatched internet-facing systems such as VPNs, and phishing. These were the top three initial access vectors across breaches in Verizon’s 2025 report.
Can antivirus stop ransomware?
It helps, but attackers often disable security tools or use legitimate admin software. Strong MFA, patching, least privilege, monitoring and offline backups are needed alongside it.
Are cloud backups enough?
Not on their own. CISA warns that encrypted local files can sync to the cloud and overwrite good copies. Keep offline or immutable backups with separate credentials and test restores.
Should I report a ransomware attack?
Yes. The FBI asks victims to report through a local field office or IC3 whether or not they pay, and many organizations also have legal reporting duties.
The bottom line
Ransomware locks or steals data and demands payment, and it shows up in a large share of breaches, especially at smaller companies. Attackers usually get in through stolen credentials, unpatched edge systems or phishing, then spend days spreading before they strike. Strong MFA, quick patching, least privilege, early detection and offline backups you’ve actually tested give you the best chance of stopping an attack or recovering without paying.
Sources: CISA #StopRansomware Guide; Help Net Security: Verizon 2025 DBIR findings; Verizon 2025 DBIR announcement; FBI IC3: ransomware victims urged to report.


