
Shashank D Shastry
Shashank is an SEO Content Specialist at G2 with over six years of experience in the B2B SaaS space. He tests and reviews tools across different software categories to create a diverse mix of content. His goal is to demystify G2 Data into clear, actionable insights for buyers navigating their software purchase decisions. When he's not working, he's likely checking out a critically acclaimed piece of pop culture or trying a new health fad.
TL;DR
- What is social engineering? A cyberattack method that manipulates people into handing over passwords, money, or access by posing as someone they trust. It targets human behavior instead of software flaws.
- How big is the problem? The 2026 Verizon DBIR found people were involved in 62% of breaches, and social engineering was the third most common breach pattern.
- How does a social engineering attack work? Attackers research a target, build trust, trigger the action they want, then disappear.
- What are the main types of social engineering attacks? Phishing (email, text, voice, QR code), pretexting, baiting, tailgating, watering hole attacks, business email compromise, and AI-assisted deepfake impersonation.
- Is phishing social engineering? Yes. Phishing is the most common form, delivered by email, text, phone call, or QR code.
- How do you spot a social engineering attack? Look for unexpected urgency, pressure to skip normal processes, lookalike sender domains, and requests for credentials, codes, or payments.
- How do you prevent social engineering? Combine regular training, phishing-resistant multi-factor authentication, call-back verification for password resets and payment changes, and a tested incident response plan.
- Which software helps? Intrusion detection and prevention, email anti-spam, endpoint protection, and security awareness training tools each cover a different stage of an attack.
Security teams spend their days hardening systems, but many of the most damaging attacks go after something no system can fully protect: the person using it. An attacker who convinces an employee to read out a one-time code, approve a wire transfer, or reset a password for a “colleague” never has to exploit a software flaw.
That’s why social engineering appears in so many incident reports, and why defense needs both people and tooling. Training and verification procedures reduce the number of people who get fooled. Controls such as intrusion detection and prevention systems (IDPS) help catch what follows when someone is, because G2’s category definition says these tools detect malware, socially engineered attacks, and other web-based threats.
This guide explains what social engineering is, how attacks unfold, the main types with real examples, and how to prevent social engineering attacks in your organization.
What is social engineering?
Social engineering is a cyberattack method that manipulates people, not software, into handing over passwords, money, or access. Attackers pose as a trusted person or organization and use urgency, fear, curiosity, or the wish to be helpful to get a target to act before they think.
These attacks work because they exploit predictable habits. People defer to authority, respond to deadlines, and want to help a colleague in trouble. Attackers study those habits, often using what a target posts on LinkedIn or a company website, and write a story that makes the requested action feel routine.
How common and costly is social engineering?
Social engineering shows up in most breach data. The figures below come from the FBI’s 2025 Internet Crime Report and IBM’s 2026 Cost of a Data Breach press release, alongside Verizon’s 2026 DBIR.
| Finding | Source |
|---|---|
| People were involved in 62% of breaches, up from 60% the year before. | Verizon 2026 DBIR |
| Phishing was the starting point for 16% of breaches. | Verizon 2026 DBIR |
| Phishing and spoofing was the most reported crime type to the FBI in 2025, with 191,561 complaints. | FBI IC3 2025 Internet Crime Report |
| One in four malicious breaches was AI-enabled, and those breaches averaged $6 million, against a global average of $4.99 million. | IBM 2026 Cost of a Data Breach Report |
Voice and text deserve as much attention as email. In Verizon’s simulations, the median successful click rate on mobile-centric channels such as voice and text messaging was 40% higher than on email.
Want to learn more about Email Anti-spam Software? Explore Email Anti-spam products.
How does a social engineering attack work?
Most attacks follow four stages, whether they run over email, phone, or in person.
- Research: The attacker gathers details on the target’s role, colleagues, vendors, and interests from social media, company sites, and leaked data. More research makes the story more believable.
- Engagement: The attacker makes contact, usually posing as a coworker, vendor, bank, or authority, and builds enough trust that the target doesn’t question the exchange.
- Attack: With trust in place, the attacker asks for the payoff: credentials, a one-time code, a payment, a file opened, or a door held open.
- Closure: The attacker ends contact quietly so the target has little reason to report it.
What are the main types of social engineering attacks?
Social engineering techniques overlap, and one attack often combines several. The table summarizes the main types, and the sections after it explain each one.
| Type | Channel | What the attacker wants |
|---|---|---|
| Phishing | Email, text, phone, QR code | Credentials, payments, or a malware install |
| Pretexting | Phone, email, chat | Information or an action, based on a made-up scenario |
| Baiting | USB drives, ads, downloads | Malware on a device |
| Tailgating | In person | Physical access to a building or room |
| Watering hole | Websites the target group visits | Malware on visitors’ devices |
| Business email compromise | A redirected wire transfer | |
| Deepfake impersonation | Video and voice calls | Payments or access approved by a fake executive |
Is phishing social engineering?
Yes. Phishing is social engineering delivered through a fraudulent message that pushes the recipient to click a link, open an attachment, or share credentials. G2’s guide to phishing covers the warning signs in detail. Common variants include:
- Spear phishing: A targeted message built from research on one person or team. Our guide to spear phishing walks through examples.
- Whaling: Spear phishing aimed at executives, usually to approve payments or share confidential data.
- Vishing: Phone-call or voicemail phishing, often posing as a bank or IT support.
- Smishing: Text messages posing as delivery alerts, bank warnings, or prizes, with a link to a fake page.
- Quishing: QR codes that lead to credential-harvesting pages.
What is pretexting?
Pretexting builds a made-up scenario so the target has a reason to share information or take an action. The attacker usually impersonates someone with authority or a legitimate need, such as a bank, an auditor, a government official, or IT support. Verizon’s 2026 report counts pretexting as the initial access vector in 6% of breaches and notes that it is increasingly used to start ransomware and extortion attacks.
Reverse social engineering is a variant. Instead of contacting the target, the attacker creates a reason for the target to make contact, such as a notice on a bulletin board that lists a fake IT help desk number.
What is baiting?
Baiting lures a target with something attractive. The bait is often physical, such as a USB drive labeled “Salary review” left in a break room, but it can also be a malicious ad or a free download. Plugging in the drive can install malware.
In a University of Illinois experiment, researchers dropped 297 USB drives across campus and estimated that between 45% and 98% were picked up and connected, with the first connection in under six minutes. Most respondents to a follow-up survey said they were trying to find the owner.
What is tailgating?
Tailgating, also called piggybacking, gets an attacker into a secured building by exploiting courtesy. The attacker follows an employee through a door, claims to have forgotten a badge, or carries an armful of boxes that makes holding the door feel natural. Some carry a fake ID card and blame a faulty reader. Training staff to politely direct visitors to reception or security is the standard fix.
What is a watering hole attack?
In a watering hole attack, the attacker infects a website that the target group visits regularly, such as an industry forum or a supplier portal. People who would never click a link in an unsolicited email may trust a site they use every day. Attackers can also filter by IP address so that only visitors from the target organization receive the malicious code.
Did you know? The name comes from predators that wait near watering holes for prey.
What is business email compromise (BEC)?
Business email compromise targets companies that make regular wire transfers. The attacker compromises or spoofs an email account, impersonates an executive or a vendor, and redirects a payment. It was the second costliest crime type in the FBI’s 2025 data, with $3,046,598,558 in reported losses across 24,768 complaints.
How does AI change social engineering?
Generative AI makes impersonation cheaper and more convincing. In 2025, the FBI’s IC3 received 22,364 complaints that referenced AI, with adjusted losses above $893 million. For BEC, IC3 noted that chat generators can produce official-sounding emails that mimic a CEO, and voice cloning can be used to request wire payments. Businesses reported more than $30 million in losses to BEC scams involving AI.
IBM’s 2026 data describes AI-enabled attacks as mostly deepfake impersonation and AI-enabled malware.
What are real examples of social engineering attacks?
Real-world social engineering attacks exploit human trust rather than software flaws to steal funds, bypass multi-factor authentication (MFA), and breach corporate networks. Notable 2026 incidents include the $6.3 billion Binance fake job interview heist, the Revolut government impersonation breach, ClickFix and ConsentFix verification scams, and IT help desk voice phishing.
| Incident / Campaign | Primary tactic | Impact |
| Binance Heist (Oct 2026) | Fake multi-round job interview with trojanized coding test | ~$6.3 billion stolen in 20 minutes |
| Revolut Breach (Sept 2026) | Fraudulent government/law enforcement data request | Exposure of customer IDs, selfies, and transaction records |
| ClickFix & ConsentFix (2026) | Fake browser/CAPTCHA troubleshooting prompts | Malware execution and OAuth session token theft |
| IT Help Desk & Cold Relic (2026) | Vishing and internal chat impersonation | Password resets and unauthorized MFA device registration |
How did a fake job interview lead to the $6.3 billion Binance heist?
In October 2026, North Korean state-sponsored hackers drained roughly $6.3 billion from cryptocurrency exchange Binance after compromising a software engineer’s work laptop through a fake multi-round job interview. Posing as recruiters on a legitimate job platform, the attackers conducted a live video interview with a purported HR representative, who even shared personal social media profiles to build trust, followed by a second interview with a fake CEO. At the end of the process, they instructed the engineer to download a coding "assignment" containing hidden zero-day malware, allowing the attackers to pivot internally and trigger unauthorized wallet withdrawals.
How did attackers trick Revolut by posing as a government agency?
In September 2026, fintech company Revolut disclosed a social engineering data breach after attackers used a compromised legitimate government-agency email domain to submit fraudulent legal data requests. Because the emails passed standard domain authentication checks, compliance staff fulfilled the requests and handed over sensitive records, including passport copies, driver's licenses, verification selfies, and transaction histories, without Revolut’s core technical systems ever being hacked.
What are "ClickFix" and "ConsentFix" verification scams?
ClickFix and ConsentFix are user-assisted social engineering attacks that trick victims into executing malicious commands or handing over login tokens under the guise of fixing a browser error or completing a CAPTCHA check. In ClickFix attacks, a fake error prompt instructs users to paste a snippet into their Windows Run dialog or terminal, silently installing info-stealing malware. In ConsentFix variants, victims are directed to a real Microsoft sign-in page and tricked into copying the resulting redirect URL back to the attacker's site, handing over an active OAuth authorization code that hijacks their authenticated session.
How do attackers abuse IT help desks and internal messaging?
Threat groups like Scattered Spider and Cold Relic breach organizations by impersonating locked-out employees or trusted colleagues to persuade IT support staff to reset passwords and register rogue MFA devices. Because help desks are trained to resolve employee issues quickly, voice phishing (vishing) calls and hijacked internal chat threads frequently succeed unless organizations enforce strict out-of-band verification, such as calling the employee back on a pre-registered corporate phone number.
How do you detect a social engineering attack?
Social engineers want you to react, not think. These warning signs show up across email, text, phone, and in-person attempts:
- A message that creates urgency, fear, or a deadline, such as a locked account or a missed payment.
- A request to skip a normal process, like paying an invoice outside the approval workflow or sharing a one-time code.
- A sender address or link domain that is close to a real one but not identical.
- A message from a senior leader or coworker asking for credentials, gift cards, or an urgent payment.
- An unexpected verification request, such as a password reset or one-time passcode, for a service you haven’t used recently.
- Prizes, refunds, or threats of legal action you weren’t expecting.
- A stranger asking for access to a building, a server room, or a device.
- A reply to a question you never asked.
Before you act on a message, ask whether the sender can prove their identity, whether you feel rushed, and whether the offer or threat is realistic.
Tip: Verify any request for money, credentials, or access through a second channel you already trust, such as a known phone number or an in-person check. Don’t use the contact details in the message itself.
How do you prevent social engineering attacks?
No single control stops social engineering, because the attacker is targeting a decision, not a vulnerability. Effective defense layers training, technical controls, verification procedures, and regular testing.
How does security awareness training reduce social engineering risk?
Regular training teaches employees to recognize manipulation and gives them a safe way to respond. Tailor it by role: finance teams need to practice payment verification, help desk staff need reset procedures, and reception staff need tailgating scripts. Because attackers use voice and text as well as email, simulations should too.
Security awareness training software such as KnowBe4 Security Awareness Training, Huntress Managed Security Awareness Training, and usecure automates phishing simulations, assigns short role-based modules, and reports click rates without heavy IT overhead, helping IT managers improve compliance and reduce risk. They make reporting easy, too: when employees can flag a suspicious message in one click, security teams see attacks earlier.
Which technical controls help prevent social engineering?
Technical controls act as a safety net when a convincing lure gets past human judgment, blocking malicious messages, domains, and account takeovers before they cause damage:
- Phishing-resistant multi-factor authentication: A stolen password is far less useful without a second factor. Choose methods that can’t be read out over the phone, such as hardware security keys or passkeys, over SMS codes or push approvals that users can be pressured into approving. G2’s guide to two-factor authentication compares the common options.
- Cloud email security and anti-spam: Email anti-spam platforms such as Proofpoint Core Email Protection, Abnormal AI, and IRONSCALES catch advanced phishing, business email compromise (BEC), and ransomware threats with high detection accuracy and low false positive rates—reducing IT admin workload for small and mid-market teams while letting buyers evaluate protection before committing to a multi-year contract.
- DNS security solutions: DNS security solutions such as DNSFilter and Cisco Umbrella protect distributed networks by blocking connections to malicious phishing and malware domains before a page ever loads.
- Disinformation and deepfake detection: Disinformation detection tools such as Blackbird.AI and TruthScan identify AI-generated content and deepfakes across social media and communications channels, providing transparent metadata and confidence scores for human security review.
- Endpoint protection and IDPS: Endpoint tools stop malware delivered through baiting, malicious attachments, and compromised websites, while intrusion detection and prevention systems (IDPS) flag abnormal behavior and block web-based threats after someone has been fooled.
Top intrusion detection and prevention software on G2
Following are leading products from G2’s Intrusion Detection and Prevention Systems (IDPS) category.
- Palo Alto Networks Next-Generation Firewalls. Best for organizations that want security policies built around application, user, and content, including encrypted traffic. G2 rating: 4.5/5
- Check Point Quantum Intrusion Prevention System (IPS). Best for teams that want IPS with automatic updates built into a Check Point next-generation firewall. G2 rating: 4.4/5
- Palo Alto Networks VM-Series Virtual Firewall. Best for protecting workloads in cloud and virtualized environments, including AWS, Azure, Google Cloud, and VMware. G2 rating: 4.3/5
- Cloud-Delivered Security Services. Best for organizations that want cloud-delivered threat prevention, including URL filtering aimed at unknown phishing sites. G2 rating: 4.4/5
- Check Point Infinity Platform. Best for teams that want one consolidated security architecture across network, endpoint, cloud, and mobile. G2 rating: 4.6/5
What policies and procedures stop attackers who get past the filters?
Some attackers get past filters and training. Procedures give employees a rule to follow when a request feels wrong:
- Verification procedures: Require a call back to a known number before changing payment details, resetting a password, or enrolling a new MFA device, and require a second approver for large transfers.
- Information-handling rules: Define what staff may share, with whom, and through which channels, so employees don’t have to guess.
- Physical security: Use badges, escort visitors, and keep used hard drives, defective USB drives, and sensitive paper in locked bins until they are properly disposed of.
- No-blame reporting: Make it easy to report a mistake quickly, because speed limits the damage.
Did you know? In 2025, the FBI’s Recovery Asset Team froze $679,013,183 of $1,163,919,846 in attempted fraudulent transfers, a 58% success rate across 3,900 incidents. The FBI’s guidance is to contact your bank and file a complaint at ic3.gov as soon as you discover a fraudulent transfer.
Why should you test your defenses against social engineering?
Run controlled tests of your defenses, including simulated phishing, vishing calls, and physical access attempts, ideally through an external team that has no inside knowledge. Pair them with penetration testing and regular vulnerability scanning, then fix the gaps they find and test again.
Frequently asked questions about social engineering
Got more questions? Here are the answers.
Q. What are the four types of social engineering?
Lists differ by source, but the four types cited most often are phishing, pretexting, baiting, and tailgating. This guide also covers watering hole attacks, business email compromise, and deepfake impersonation.
Q. What is an example of social engineering?
A common example is a text message that claims to come from a delivery company, says a parcel is on hold, and links to a fake page that asks for card details. At work, a typical example is a caller who poses as IT support and asks you to read out a one-time code.
Q. Who carries out social engineering attacks?
Several groups use social engineering techniques:
- Cybercriminals and fraud rings: Seek money through phishing, BEC, and ransomware.
- State-backed actors and spies: Gather information on organizations for strategic or financial advantage.
- Identity thieves: Steal personal data such as names, addresses, and Social Security numbers to sell or misuse.
- Malicious insiders: Current or former employees who misuse access or trick colleagues, often out of grievance or for payment.
- Penetration testers: Authorized professionals who run controlled attacks to test how well people and processes hold up.
Q. What is the best defense against social engineering?
No single control works alone. The strongest setup combines regular training, phishing-resistant MFA, call-back verification for payments and resets, monitoring, and tested incident response. Verification habits are the control that still works when a message or call looks convincing.
Q. Can AI make social engineering harder to spot?
Yes. AI removes the spelling mistakes and awkward phrasing that used to give phishing away, and it can clone voices and faces. For high-risk requests, treat voice and video as unverified and confirm payments and access changes through a second channel, even when the request comes from a call with familiar faces.
Q. What should you do if you fall for a social engineering attack?
Act quickly and in this order:
- Report it to your IT or security team immediately, even if you’re embarrassed.
- Change the affected passwords and ask IT to revoke active sessions and MFA devices.
- If money moved, call your bank to request a recall and file a complaint at ic3.gov.
- Save the messages, call logs, and any attachments as evidence.
- Follow your incident response plan to contain and investigate the incident.
Q. What are the best email security and anti-spam tools for stopping phishing and BEC?
According to G2 user reviews, top-rated platforms in the Email Anti-Spam category include Proofpoint Core Email Protection for large enterprises, Abnormal AI for behavioral detection with low false positives, and IRONSCALES or Check Point Email Security for mid-market and small IT teams looking to reduce admin workload. Pairing email filtering with DNS security tools such as DNSFilter or Cisco Umbrella adds a second layer that blocks malicious links before pages load.
Q. Which tools help teams spot deepfakes and train employees against social engineering?
For detecting AI-generated media and deepfakes, security teams use disinformation detection tools such as Blackbird.AI and TruthScan, which surface metadata and confidence scores for human review. To reduce employee susceptibility across email, voice, and SMS lures without heavy IT overhead, top-rated security awareness training platforms on G2 include KnowBe4, Huntress Managed Security Awareness Training, and usecure.
Making verification a habit
Social engineering will keep adapting, from help desk calls to deepfake video meetings, because manipulating a person is often easier than breaking a system. The pattern across the examples above is consistent: the attacker creates urgency, borrows authority, and hopes nobody checks.
Teams that make verification routine, back it with phishing-resistant MFA and monitoring, and rehearse their response limit the damage when someone is fooled.
Ready to stop social engineering before it starts? See how the leading security awareness training tools stack up in our complete comparison guide.
This was originally written by Sagar Joshi in 2021. It has been updated with new information.
