Skip to content
Let’s talk
Manos tecleando en un portátil en una habitación a oscuras

Types of malware: 9 categories and how to reduce the risk

An employee opens a PDF from a regular supplier. Nothing seems odd. Three weeks later, the company discovers that someone has been inside for twenty-one days reading emails, copying databases and waiting for the moment to demand a ransom. It wasn’t just any virus. It was a remote access trojan — a RAT in the jargon — disguised as an invoice.

Types of malware can combine techniques and affect devices, identities and data. Antivirus, EDR, network controls and monitoring provide different and complementary capabilities. No tool guarantees detecting or stopping every attack; configuration, coverage and response matter too.

This guide presents nine categories of malware, signs worth investigating and measures that can reduce their impact.

What malware is (and why the word is confusing)

Malware is short for malicious software: any program, script or code designed to damage, steal, spy on or take control of a device without the owner’s permission. It’s an umbrella term. It includes viruses, worms, trojans, ransomware, spyware, keyloggers, rootkits, bots, infostealers, malicious adware and fileless malware.

The problem with the plain term “malware” is that it doesn’t tell you how it got in or how it behaves. And without that, you can’t defend yourself. Calling ransomware or an infostealer “malware” is like calling both flu and cancer “illness”: technically correct, operationally useless.

The 9 types of malware you should know

The list is a practical selection of categories, not a statistical ranking by impact. Some overlap and a single attack can use several.

Ransomware

It can encrypt files and demand a ransom. Some campaigns combine encryption and data theft to increase pressure, and some use affiliates under a service model. Entry routes and the time to impact vary. Prevention, detection, isolated backups and recovery should form a joint plan.

Trojans (including RATs)

Malware that appears to be legitimate software or content and can open back doors, steal data or install other payloads. RATs allow unauthorised remote access. Detection depends on the variant, configuration and controls; there’s no universal percentage for comparing all antivirus products with all EDRs.

Infostealers

Malware specialised in stealing credentials, session cookies and other browser data. A stolen session can allow access without repeating authentication while it remains valid and the service doesn’t block it. This doesn’t mean every MFA mechanism can be bypassed in any situation. MITRE describes the misuse of session cookies.

Spyware

Spy software that collects information without permission: what you type, which websites you visit, which documents you open, which calls you make. It ranges from corporate-grade spyware (Pegasus and similar, government-level offensive cyber) to low-cost commercial variants sold to companies to “monitor employees”. The difference from an infostealer is that spyware watches continuously rather than doing a quick extraction. Detectable through anomalous network behaviour (constant uploads to unusual IPs).

Keyloggers

Keyloggers record keystrokes and can be software or physical devices. Inspecting the device and monitoring can help detect them, but an EDR doesn’t guarantee identifying every keylogger, especially a physical device outside its telemetry.

Worms

Worms replicate and can spread without any new action by the user. Patching and segmenting reduce opportunities to spread. Separating VLANs isn’t enough on its own: traffic between segments has to be controlled and permissions, routes and exceptions reviewed.

Computer viruses

A virus inserts malicious code into other files and can replicate when they are run. Antivirus products detect many known variants and some new ones, but no product guarantees identifying them all. The risk depends on the environment and exposure.

Rootkits

Rootkits aim to hide malicious activity and can affect system components with elevated privileges. They make detection harder; they aren’t necessarily invisible. Investigation may require tools and a trusted environment, and in some cases rebuilding the system.

Fileless malware

The term fileless covers techniques that run parts of the attack in memory or through legitimate tools. Some phases may use files. Modern antivirus can analyse scripts, memory and behaviour; Microsoft documents these capabilities through AMSI (source). Restrictions should be adjusted to operational needs and controls.

Bots, botnets, malicious adware, cryptojacking, wipers

There are more categories that don’t fit in a useful list. Bots and botnets are infrastructure (an infected machine that joins a network controlled by an attacker, usually for DDoS attacks or sending spam). Malicious adware and cryptojacking usually come in through untrustworthy installers. Wipers erase data without demanding a ransom — they’re used for sabotage, not business. They’re real, but they’re not usually the first operational concern for an average company today.

How malware gets into a company

These six routes are useful for organising an exposure review. They don’t represent a universal percentage of infections or an exhaustive list.

Entry routeWhat it exploitsLayer that stops it
Email phishingMalicious attachments or links. A user who clicksAdvanced filtering + training + MFA
Exposed RDP / VPNWeak or stolen credentials, no MFAMandatory MFA, VPN without direct RDP, minimal exposure
Unpatched vulnerabilityExchange, Fortinet, Cisco, browser, CMS pluginVulnerability management + prioritised patching (EPSS)
Malicious website or drive-byA user visiting a compromised site or malvertisingDNS filtering + EDR + up-to-date browser
Pirated software / cracksTrojanised installers for “free” programsControlled software policy + legal licences
Supply chainCompromised supplier, malicious signed updateSIEM monitoring + principle of least privilege

Both phishing and exposed remote services can trigger serious incidents. The scope depends on the access obtained and on subsequent controls; it can’t be inferred from the entry route alone.

Signs there’s malware inside your network

The following signs justify investigating, but they can also have legitimate causes or be technical faults. They need correlating with context and other evidence before concluding that malware is present.

Connections to unusual IPs

Connections to unexpected or newly registered destinations may deserve a review. An IP’s geolocation on its own doesn’t identify an attack. Network logs and correlating them help the investigation.

Suspicious PowerShell or scripts

Encoded parameters, unexpected commands and new tasks can be useful indicators. They also appear in legitimate administration: review the origin, user, process and purpose before attributing them to malware.

Accounts with anomalous usage

Logins outside the usual pattern or unexpected privilege changes need context. Detection depends on identity sources, rules and the coverage of the tools available.

Processes you don’t recognise

Binaries running from %AppData%, %Temp% or odd paths, newly created services, child processes of Word or Excel launching cmd.exe. All of this is detectable and should trigger an alert.

Exfiltration: large uploads

An unusual transfer may indicate exfiltration, but also a backup or another legitimate operation. Investigate the process, destination and authorisation; on its own it doesn’t prove that ransomware is about to encrypt data.

Antivirus or EDR disabled

Loss of telemetry from an antivirus or EDR needs priority investigation. It may be due to tampering, connectivity problems or agent faults. Assess the risk and follow the response procedure without treating an infection as confirmed on that signal alone.

Outside emails warning about spam from you

Suspicious messages may come from a compromised account or from sender spoofing. Review headers, sessions and logs; if compromise is confirmed, apply session revocation and the planned recovery measures.

How to defend your company against malware

No single layer stops everything. Combining controls reduces exposure and makes response easier, but the residual risk depends on how they’re implemented and on the environment. Prioritise measures based on an assessment of your case.

Complementary antivirus and EDR/XDR

Antivirus may include signatures, heuristics and behavioural analysis. EDR adds telemetry, investigation and response on endpoints; the specific capabilities depend on the product and licence. There’s no fixed detection rate that universally distinguishes the two categories. Microsoft documents behavioural and memory analysis in its antivirus.

Mandatory MFA everywhere

Prioritise MFA on email, remote access and privileged accounts, preferably phishing-resistant where feasible. Record technical exceptions and apply compensating measures. The absence of MFA increases exposure, but it doesn’t on its own explain a fixed percentage of incidents or guarantee that ransomware will happen.

Immutable, verified backup

Keep isolated or immutable copies with separate credentials and test restores according to criticality and recovery objectives. Check the real scope of immutability and retention; a backup that exists but hasn’t been tested doesn’t guarantee a successful recovery. We do it in business backup.

Network segmentation

VLANs separating IT, OT, production, servers, guests. An incident that starts on an admin PC shouldn’t be able to touch the file server or the ERP without passing through an internal firewall.

Vulnerability management

Inventory, vulnerability review and patching prioritised by asset criticality, exposure, known exploitation and severity. EPSS and CVSS provide information, but they don’t replace context or the vendor’s recommendations.

Training and simulations

Authorised training and simulations with clear goals, regular reviews and comparable metrics. Use the results to improve processes and training, without promising a percentage reduction in clicks or a universal timeframe.

24/7 monitoring (SOC or MDR)

Assign owners to review and handle alerts, including the out-of-hours periods the risk requires. If you contract a 24/7 SOC or MDR, confirm the coverage, response times, exclusions and authorised actions in writing. More in managed SOC for SMEs.

Primary sources: CISA StopRansomware · NIST SP 800-61r3 · Microsoft Defender for Business.

Types of malware FAQs

What’s the difference between a virus and malware?

Malware is the umbrella term for any malicious software. A virus is a specific type of malware: one that attaches to legitimate files and replicates when they are run. Every virus is malware; not all malware is a virus. In fact, most of the malware we see in SMEs today isn’t viruses: it’s trojans, infostealers and ransomware.

Does my antivirus protect me against ransomware?

An up-to-date, well-configured antivirus can detect or block ransomware, but not every attack. Combine it with access controls, protected backups, segmentation and investigation and response capability according to the risk. EDR can add capabilities; it doesn’t guarantee total protection either.

What is an infostealer and why is it so dangerous?

An infostealer steals data such as credentials and session cookies. If a stolen session is still valid and can be reused, the attacker could gain access without repeating authentication. The outcome depends on the service’s controls; it doesn’t mean any MFA always fails.

How does malware get in without anyone clicking?

Through unpatched vulnerabilities in exposed services: Exchange, SSL VPN, management panels. The attacker doesn’t need anyone to click; they exploit the software flaw to run code. Also through the supply chain (a legitimate but compromised update) or credentials stolen previously in another incident.

Is it enough to have Microsoft Defender on our computers?

You need to distinguish Microsoft Defender Antivirus from the products and plans that add EDR capabilities. Defender for Business includes EDR; in Defender for Endpoint the scope depends on the plan. Review the licence, configuration, coverage and who handles the alerts before concluding whether the solution is enough for your risk.

How long does an attacker take from getting in to encrypting?

There’s no fixed window of 3 to 21 days that applies to every attack. Encryption or data theft can happen quickly or after a prolonged stay. Prepare detection and response without assuming there will always be several days to act.

How do I know whether my company is already infected without us knowing?

Review the anomalous signs and relate them to logs and context. None of them on its own confirms an infection. Depending on severity, a compromise assessment or a response team may be needed; keep the relevant evidence and follow the incident plan.

Does Mac malware exist? We were told it doesn’t…

Yes. macOS can also be affected by malware, including software that steals information. Keep the system up to date, control applications and permissions and assess the device’s protection and monitoring needs. The choice of EDR depends on the risk and how the environment is managed, not just the operating system.

Etiquetas

Keep reading

See the blog

Would you rather we explained it for your case?

If a guide has left you with a specific question about your company, tell us about it in a free 30-minute call. We’ll tell you where to start, no strings attached.

Book a free 30-minute call

+34 871 030 201 · Carrer de Can Martí Feliu, 4, Palma

Cookies

We use Google Analytics and Google Ads cookies to see how people find us and which enquiries the website brings in. They are only switched on if you accept. Cookie policy