Phishing is a cyberattack in which criminals pose as a trusted company, colleague, or service to trick people into giving up passwords, payment details, or account access. It arrives by email, text message, phone call, QR code, social media, and paid ads. In the second quarter of 2026 alone, the Anti-Phishing Working Group (APWG) recorded 1,069,681 phishing attacks.
Most phishing defenses act when a message lands in an inbox. By then, the attack is already built. Attackers spend days or weeks registering domains, setting up hosting, and testing campaigns before the first lure goes out. This guide covers the full attack, from that preparation stage to recovery.
What Is Phishing?
Phishing is a form of cybercrime that combines social engineering with technical deception to steal personal data and financial account credentials. Attackers impersonate a legitimate organization through fake emails, messages, websites, and domain names, then lead victims to counterfeit pages or malware that capture logins and payment information.
Phishing is fraud that exploits trust. Instead of breaking through software flaws, it targets people. The FBI defines it as unsolicited emails, texts, and phone calls that claim to come from a legitimate company and ask for personal, financial, or login details. The goal is usually stolen credentials, stolen money, system access, or a foothold for malware.
No organization is too small to be impersonated. In Q2 2026, APWG identified 941 unique brands in phishing reports, and half of them were attacked five times or fewer. Phishers keep building page templates for lesser-known and regional targets, so a regional bank or mid-sized manufacturer is as exposed as a global brand.
Related Read: History of Phishing Attacks
Phishing vs. Spam vs. Spoofing
| Phishing | Spam | Spoofing | |
|---|---|---|---|
| What it is | A deceptive message designed to steal data, money, or access | Unsolicited bulk messages, usually advertising | Faking a sender’s identity, such as an email address, phone number, or domain |
| Attacker’s goal | Credentials, payments, or system access | Clicks, sales, or traffic | Making a message look trustworthy |
| How they relate | Usually relies on spoofing | Can carry phishing, but most spam is not phishing | A technique used inside phishing and fraud |
The overlap is why the FBI tracks phishing and spoofing as one crime category.
Related Read: What Is a Spoofing Attack? What Financial Institutions Need to Know.
How a Phishing Attack Works, From Domain Registration to Payout
A phishing attack moves through five stages. Most security tools only see stages 3 and 4, when a message is delivered and a victim clicks. The earlier stages are where attackers do most of their work, and where defenders have the most room to act.
Stage 1: Attackers Build the Infrastructure
Before sending a single message, attackers need somewhere to send victims. That usually means:
- Registering lookalike domains with swapped letters, extra words, or different extensions, a technique called typosquatting or URL hijacking.
- Choosing hosting, often behind services that hide the origin server.
- Installing a phishing kit, a ready-made copy of a login or checkout page, often rented as a subscription.
APWG’s Q2 2026 data shows Cloudflare’s proxy hosting remained popular with phishers, followed by RouterHosting and Amazon. For business email compromise campaigns, NameCheap was the registrar used most often, followed by NameSilo and Dynadot.
One phishing site can sit behind thousands of customized URLs, so blocking a single link does little while the domain and hosting stay live.
Related Read: How Can You Tell if a Domain Is Malicious?
Stage 2: Attackers Choose Targets and Write the Lure
Some campaigns go to millions of addresses. Others are researched and aimed at one person, such as a finance manager. The lure plays on urgency, fear, authority, or reward: an unpaid invoice, a locked account, a message from the CEO.
Lures are built to slip past filters. In Q2 2026, APWG reported more two-step phishing, where a normal-looking business inquiry comes first and the phishing link only follows after the victim replies. Generative AI also removes the spelling and grammar mistakes people were once taught to spot.
Stage 3: Attackers Deliver the Lure
Email is still the most common channel, but phishing now arrives by SMS, phone calls, QR codes, social media messages, and paid search and social ads. Phone and text are growing fastest: in Q2 2026, vishing rose 20% and smishing rose 40% on the previous quarter. Threat activity across Meta Ads surged 571% in the same period.
Stage 4: Attackers Capture What They Came For
Once a victim engages, the attack collects usernames and passwords, session tokens, payments, or installs malware. Session theft is now common against organizations that use multi-factor authentication (MFA).
In April 2026, Microsoft tracked a campaign that targeted more than 35,000 users across over 13,000 organizations in 26 countries. Its final stage was an adversary-in-the-middle (AiTM) flow that captured authentication tokens and bypassed MFA that was not phishing-resistant.
Stage 5: Attackers Cash Out and Repeat
Stolen access becomes money. Attackers drain accounts, redirect payments, resell credentials, or use a compromised mailbox to phish the victim’s colleagues and customers. In Q2 2026, gift cards were the most common business email compromise cash-out method, and the average wire transfer requested was $61,732, up 45% on the previous quarter.
When one campaign ends, the kits, hosting, and techniques move on to the next. That is why disrupting the infrastructure matters more than blocking one link after it has reached inboxes.
12 Types of Phishing Attacks
Phishing types are named after the channel they use or the target they go after. Many campaigns combine several: a smishing text might lead to a lookalike domain running an AiTM phishing kit.
| Type | Main channel | Typical target | Attacker’s goal |
|---|---|---|---|
| Email phishing | Anyone | Credentials, malware | |
| Spear phishing | Email, messaging | Specific individuals | Credentials, access |
| Whaling | Email, phone | Senior executives | Payments, sensitive data |
| Business email compromise | Finance and payroll staff | Wire transfers, gift cards | |
| Clone phishing | Recipients of a real email | Credentials, malware | |
| Smishing | SMS, messaging apps | Consumers, employees | Credentials, payment details |
| Vishing | Phone, voice calls | Consumers, help desks | Account access, payments |
| Quishing | QR codes | Consumers, employees | Credentials, payment details |
| Social media and ad phishing | Social platforms, paid ads | Customers, job seekers | Payments, personal data |
| AiTM phishing | Fake login pages | Users protected by MFA | Session tokens |
| AI-generated phishing | All channels | Anyone | Higher success rates |
| Lookalike domain phishing | Fake websites | A brand’s customers | Credentials, payments |
Real-World Phishing Examples From 2026
A BEC gang using fake coaching invoices: APWG named one group as the most active BEC operation in Q2 2026. It uses fake executive coaching invoices and a spoofed reply chain between the coaching firm and an executive to make its requests look credible, and controls a large network of mule accounts, most held at US banks.
A compliance-themed AiTM campaign: Between April 14 and 16, 2026, Microsoft tracked campaigns that targeted more than 35,000 users across over 13,000 organizations in 26 countries, ending in an AiTM flow that captured authentication tokens. The lures posed as internal code-of-conduct notices, a theme that pushes employees to act fast.
Malicious ads on social platforms: Threats grew on LinkedIn by 106%, YouTube by 118%, and Facebook by 88% from Q1 to Q2 2026.
Which Industries Attackers Target Most
Every sector is targeted, but attackers go where the money, logins, and customer trust are. On social media, APWG found finance remained the most targeted sector in Q2 2026 at 27.1% of threats.
1. Financial Services: Phishing and Brand Impersonation
Banks, lenders, insurers, and payment providers are among the most impersonated brands. Attackers register lookalike banking domains, copy online banking logins, and send fake fraud alerts by text. The cost lands on customers through account takeover, and on the institution through reimbursements and lost trust.
Related Read: The Top Four Things Bankers Should Know About Phishing
2. Manufacturing: Supplier and Partner Impersonation
Long supplier chains and frequent invoices make manufacturers a natural target for BEC, where a fake supplier email changes bank details on a real invoice. Phishing can also be the first step toward malware that disrupts production.
Learn More: Protection for the manufacturing sector
3. Retail: Fake Stores and Fraudulent Promotions
Retailers face counterfeit shopping sites, fake discount codes, and campaigns timed to sales events. These attacks steal card details and damage the brand even when the retailer’s own systems were never breached.
Related Read: Cybersecurity and brand protection for retail.
4. Media and Entertainment: Audience and Account Targeting
Streaming services, publishers, and entertainment brands attract fake subscription renewal pages, impersonated artist and studio accounts, and giveaway scams.
Learn More: Media and entertainment brands.
How to Recognize a Phishing Attempt
Phishing is harder to spot than it used to be, but most attempts still share a few warning signs. Individuals can check the message in front of them. Organizations can watch for a campaign being prepared.
Red Flags for Individuals
- The message creates urgency, such as an account suspension or an invoice due within hours.
- It asks you to log in, confirm payment details, or approve an MFA request you didn’t start.
- The sender’s address or link domain is slightly different from the real one.
- A colleague or executive asks for a payment, gift cards, or new bank details through an unusual channel.
- A QR code asks you to sign in or pay.
- A caller claiming to be your bank or IT team asks for a code, password, or remote access.
When in doubt, don’t use the link, number, or reply address in the message. Go to the organization’s website or app directly, or call a number you already trust.
Warning Signs Organizations Can See Earlier
Security teams can spot many campaigns before an employee or customer sees a message. Signals include:
- New domain registrations resembling the organization’s brand or product names.
- Lookalike sites copying its login or checkout pages.
- Paid search or social ads using its name or logo without permission.
- Fake social profiles posing as its support staff or executives.
- Unusual sign-ins, new mailbox forwarding rules, or MFA registrations from unfamiliar devices.
Related Read: This Is How You Can Spot Fake Domains.
How to Prevent Phishing: A Layered Approach
No single control stops phishing. Effective programs combine technical controls, payment processes, employee awareness, and visibility into attacker infrastructure before campaigns go live.
Authenticate Your Email Domains
SPF, DKIM, and DMARC let receiving servers check whether email claiming to come from your domain is genuine. A DMARC policy set to quarantine or reject makes exact-domain spoofing much harder. These controls don’t stop lookalike domains, so they are one layer among several.
Use Phishing-Resistant MFA
Codes and push approvals can still be phished or intercepted. CISA urges all organizations to adopt phishing-resistant MFA, naming FIDO/WebAuthn and PKI-based methods such as smart cards. Where that isn’t yet possible, it recommends number matching for push-based MFA as an interim step. See the CISA fact sheet on implementing phishing-resistant MFA.
Build Verification Into Payment Processes
BEC succeeds when one email can change where money goes. Require a callback to a known number before changing supplier bank details, use dual approval for large transfers, and treat any request for gift cards as a red flag.
Train Employees on Current Tactics
Training should cover smishing, vishing, QR codes, AI-written messages, and fake MFA prompts alongside email. Make reporting quick and blame-free, so one employee’s report can protect everyone else.
Monitor and Take Down Brand Impersonation
Organizations need visibility into lookalike domains, fake sites, fraudulent ads, and impersonating social accounts, with fast takedowns for each. Reactive monitoring only catches these once they are live and may already have reached customers.
Lock Down Help Desk Resets
Vishing callers often target the help desk, not the user. A convincing caller asks for a password reset or a new MFA device and walks away with the account. Require identity checks that a caller cannot answer from social media or a data breach, such as a callback to the number on file or approval from the user’s manager, before any reset.
Disrupt Attacker Infrastructure Before Campaigns Launch
Predictive cybersecurity acts in Stage 1, while attackers are still preparing. Bfore’s PreCrime™ platform analyzes behavioral signals, infrastructure creation patterns, registration activity, and hosting behaviors to uncover threats before activation. According to Bfore, PreCrime™ identifies malicious infrastructure an average of 18 days before conventional threat intelligence sources.
What to Do After a Phishing Attack
Speed matters most. The sooner a password is changed or a session revoked, the less time an attacker has to use it.
If You Clicked a Link or Entered a Password
- Change the password for that account, and for any other account that uses it.
- Sign out of all sessions, or ask IT to revoke them, to cut off stolen session tokens.
- Turn on MFA, ideally a phishing-resistant method.
- If you entered card or bank details, contact your bank to block the card.
- If you opened a file, disconnect the device and have it checked for malware.
- Report the incident to your IT or security team, even if nothing seems wrong.
If Your Organization or Brand Was Targeted
- Contain the account: Reset credentials, revoke sessions, and check for new forwarding rules or MFA devices added by the attacker.
- Take down the infrastructure: Request removal of phishing sites, lookalike domains, and fake social accounts. If you’re under attack now, request a takedown. Takedowns have grown harder as attackers move between registrars and hosts.
- Warn the people at risk: Tell employees, and customers where needed, what the fake messages look like.
- Report it: In the US, file a report with the FBI’s Internet Crime Complaint Center, and forward phishing emails to APWG at [email protected].




