Spear phishing, whaling and business email compromise (BEC) are often lumped together as "targeted phishing", yet they aim at different people, carry different payloads and are stopped by different controls. A secure email gateway that catches a weaponised attachment does nothing against a polite, malware-free payment request sent from a supplier's real, hijacked mailbox. Treating the three as one problem leads to awareness training aimed at the wrong staff, email filters tuned for the wrong signals and finance processes that trust the channel instead of the requester. This guide compares the three side by side: the threat model for each, the controls that actually break each one, the detection signals that differ, and the failure patterns that recur. It is for security architects, email and identity administrators, SOC leads and the finance and executive assistants' teams who sit on the receiving end. For phishing fundamentals, see the Phishing Guide.
Key takeaways
- They differ by goal, not just by target. Spear phishing usually wants code execution or credentials, whaling wants an executive's authority or access, and BEC wants a payment or sensitive data, often with no link or attachment at all.
- The relationship is nested, not exclusive. Whaling is spear phishing aimed at senior leaders, and spear phishing or whaling is often how attackers gain the mailbox they later use for BEC.
- Payload controls stop spear phishing, process controls stop BEC. Sandboxing, link rewriting and phishing-resistant MFA break the first; out-of-band verification and dual approval break the last.
- DMARC helps against two of the three only partly. An enforcing policy on your own domain stops exact-domain spoofing, but not lookalike domains or messages from a genuinely compromised supplier mailbox.
- Speed decides BEC losses. The FBI reports that its Recovery Asset Team froze funds in 66% of the Financial Fraud Kill Chain cases it ran in 2024, and that the process works best when complaints are filed quickly.
Definitions and how the three relate
All three are forms of social engineering that rely on the target trusting a sender. MITRE ATT&CK models the delivery under T1566 Phishing (with sub-techniques T1566.001 Spearphishing Attachment, T1566.002 Spearphishing Link, T1566.003 Spearphishing via Service and T1566.004 Spearphishing Voice) and the deception itself under T1684.001 Impersonation, whose description explicitly names business email compromise as a case where impersonation is used to deceive victims into sending money. The money movement is T1657 Financial Theft.
- Spear phishing: a phishing message tailored to a specific person or small group, using researched details (name, role, current project, supplier) to raise the chance the target clicks, opens or enters credentials. The payload is usually technical: a malicious attachment, a credential harvesting link or an adversary in the middle (AiTM) proxy. See the AiTM Phishing Guide.
- Whaling: the UK National Cyber Security Centre (NCSC) describes whaling as highly targeted phishing aimed at senior executives, disguised as a legitimate email, and designed to get the victim to take a secondary action such as starting a wire transfer. The target is chosen for authority: an executive who can approve payments, override process or hand over strategic information.
- Business email compromise: the FBI's Internet Crime Complaint Center (IC3) defines BEC as a scam targeting businesses and individuals who perform legitimate transfer-of-funds requests, typically by compromising real business or personal email accounts through social engineering or computer intrusion. Its 2024 report adds that the same fraud now uses phone numbers and virtual meeting applications as well as email.
The practical consequence: spear phishing and whaling describe who is targeted and how precisely, while BEC describes what the attacker is trying to achieve and usually whose identity they borrow. One campaign can be all three: a spear phishing link steals a chief financial officer's mailbox credentials (whaling), and that mailbox then sends payment instructions to the accounts payable team (BEC).
Threat model compared
| Dimension | Spear phishing | Whaling | Business email compromise |
|---|---|---|---|
| Typical target | Named individuals with useful access: developers, administrators, help desk, HR | Executives, board members, their assistants and chiefs of staff | Finance, accounts payable, payroll, procurement, and the suppliers and customers who pay you |
| Who is impersonated | A service, colleague or vendor the target trusts | A peer executive, regulator, law firm, investor or board member | An executive, a supplier, a customer, a lawyer, or the employee whose payroll is being redirected |
| Payload | Attachment, link, credential prompt, OAuth consent, QR code | Link or document, or a pure request with no payload | Usually none: text that asks for a payment, a bank detail change, gift cards or data |
| Attacker goal | Initial access: code execution, credentials, session tokens | Access to an executive's mailbox, authority to bypass process, or strategic data | Money movement or sensitive data (W-2 or payroll records, customer lists) |
| Typical sending infrastructure | Newly registered or lookalike domains, compromised third-party sites, file sharing services | Lookalike domains, free webmail accounts styled as the executive, spoofed display names | Genuinely compromised mailboxes (yours or a supplier's), lookalike domains, abused cloud email services |
| ATT&CK techniques | T1566.001, T1566.002, T1566.003, T1598 Phishing for Information | T1566, T1591.004 Identify Roles, T1684.001 Impersonation | T1586.002 Compromise Accounts: Email Accounts, T1114.003 Email Forwarding Rule, T1564.008 Email Hiding Rules, T1534 Internal Spearphishing, T1657 Financial Theft |
| What filters see | A detectable artefact: file, URL, sender reputation | Sometimes an artefact, often only an unusual sender and pressure language | Often nothing malicious: a clean, authenticated message from a known correspondent |
Spear phishing: the access play
The attacker researches a target (T1589.002 Gather Victim Identity Information: Email Addresses, public code repositories, conference talks, job adverts) and builds a lure that fits that person's work. The architecture is technical: a delivery domain, a payload host, and a collection back end such as an AiTM proxy that captures the session cookie after MFA. The decisive control point is therefore authentication and execution: if the stolen password or token cannot be replayed, and the attachment cannot run, the campaign fails even when the user clicks.
Whaling: the authority play
Whaling uses the same delivery mechanics but chooses targets for authority rather than technical access. The NCSC notes that attackers draw on personal details from professional and personal social media, impersonate trusted partners such as law firms, and may follow the email with a phone call to add legitimacy. Two distinct risks follow: an executive whose account is compromised becomes the most convincing BEC sender in the organisation, and an executive who acts on a fake request can override the very controls that would otherwise stop the fraud. See the Executive Protection and Digital Footprint Guide.
BEC: the process play
BEC attacks the business process, not the endpoint. The common patterns are executive impersonation ("CEO fraud"), supplier invoice and bank detail change fraud, payroll diversion, and data theft requests. The attacker often sits silently in a compromised mailbox, sets rules to hide replies, waits for a real invoice thread, and then inserts new bank details. The IC3 2024 Internet Crime Report recorded 21,442 BEC complaints with adjusted losses of $2,770,151,146, against 193,407 phishing and spoofing complaints, which shows the asymmetry: phishing is the high-volume entry point, BEC is where the money is lost. For figures over time, see the Business Email Compromise and Fraud Statistics Guide.
Controls that break each attack
Controls fall into three layers: email and domain controls, identity controls, and business process controls. Each attack type is weakest against a different layer, which is why a single layer always leaves a gap.
| Control | Spear phishing | Whaling | BEC |
|---|---|---|---|
| DMARC at p=reject on your own domains (RFC 9989) | Stops exact-domain spoofing of internal senders | Stops spoofing of the executive's real address | Stops exact-domain spoofing only; no effect on lookalike domains or compromised real mailboxes |
| Attachment sandboxing and link rewriting or time-of-click checks | Primary control | Useful where a payload exists | Little effect on payload-free requests |
| Lookalike domain and display-name impersonation detection | Helpful | Primary technical control for executive names | Primary technical control for supplier and executive impersonation |
| External sender tagging and first-contact warnings | Helpful | Helpful, especially on mobile clients | Helpful against new lookalike senders; no effect when the real supplier mailbox is compromised |
| Phishing-resistant MFA (FIDO2 passkeys, smart cards) | Primary control against credential and AiTM theft | Primary control for executive accounts | Prevents the mailbox takeover that powers internal BEC |
| Mailbox rule, forwarding and delegate monitoring | Post-compromise detection | Post-compromise detection on executive mailboxes | Primary detection of attacker persistence |
| Out-of-band verification of payment and bank detail changes | Not applicable | Stops executive-instructed payments | Primary control: the step that breaks the fraud |
| Dual approval and payee change cooling-off period | Not applicable | Prevents one executive overriding process | Primary control |
| Executive digital footprint reduction | Limited | Primary preventive control | Reduces the material for convincing pretexts |
| Role-based awareness training | All staff, focused on links, attachments and consent prompts | Executives and assistants, focused on pretexts and secrecy requests | Finance and procurement, focused on process adherence over sender trust |
Email authentication is covered in depth in the DMARC, SPF and DKIM Guide. RFC 9989, published in May 2026 as a Proposed Standard, now obsoletes RFC 7489 and RFC 9091; its policy values remain none, quarantine and reject, with sp and np tags for subdomains and non-existent subdomains. Publishing np=reject closes a gap attackers use when they send from invented subdomains of your real domain.
Process controls for BEC and whaling
- Verify through a channel you already hold: call back a number from the supplier master file or the internal directory, never one supplied in the request. The IC3 public service announcement of September 2024 advises verifying account information changes through a secondary channel.
- Treat the requester's seniority as irrelevant: write into the payment policy that no executive can waive verification, and say so publicly to finance staff, so a junior employee checking with head office is doing their job rather than questioning authority.
- Separate the request channel from the approval channel: payments are approved in the finance system with strong authentication, not in email, chat or a call.
- Add friction for first-time and changed payees: a hold period, a second approver and a test payment for new bank details.
- Template the rules: see the Payment Verification Policy Template and Business Email Compromise Awareness for Finance Teams Guide.
Detection signals that differ
The SOC needs different detections for each type, because the observable evidence lives in different places: the message for spear phishing, the sender identity for whaling, and mailbox and payment behaviour for BEC.
| Signal | Where to find it | Strongest for |
|---|---|---|
| Detonation verdicts, malicious URL clicks, blocked attachments | Email security logs, secure web gateway, EDR | Spear phishing |
| Sign-in from a new device or infrastructure followed by token reuse from a second location | Identity provider sign-in logs | Spear phishing (AiTM), whaling account takeover |
| Display name matches an executive but address is external or webmail | Email security impersonation rules, message trace | Whaling, executive-impersonation BEC |
| Sender domain registered recently or one character away from a known supplier | Domain monitoring, threat intelligence, message headers | Whaling, supplier BEC |
| Reply-To differs from From, or replies route to an external address | Message headers | BEC |
| New inbox rule that moves, deletes or marks read messages containing "invoice", "payment" or "bank" | Mailbox audit log | BEC (T1564.008) |
| New external forwarding rule or delegate added | Mailbox audit log | BEC (T1114.003, T1098.002) |
| Bank detail change requested within a live invoice thread | Accounts payable system, supplier master change log | Supplier BEC |
| Payment to a first-time beneficiary account in an unexpected country | Treasury and banking platform | BEC |
Abused cloud email infrastructure is a growing variant: in the TruffleNet campaign, a stolen cloud access key with email sending rights was used to register sending identities for hijacked domains, so fraudulent vendor invoices passed authentication checks. Authentication results alone therefore cannot clear a payment request; see the Email Identity and BEC Guide for mailbox and application access controls.
Worked scenario: one campaign, three stages
A mid-sized manufacturer receives a series of events over three weeks. Walking through them shows where each control applies.
- Week 1, spear phishing: a buyer in procurement receives a "shared drawing" link from a lookalike of a real supplier domain. The link leads to an AiTM page. Control check: phishing-resistant MFA on the buyer's account would make the captured session useless; time-of-click URL analysis might block the page; a first-contact warning would flag the new domain.
- Week 2, whaling: using the buyer's mailbox, the attacker learns the finance director's name and sends her a "board paper" link from the internal account (T1534 Internal Spearphishing). She signs in from a hotel. Control check: an internal sender bypasses many external filters, so identity signals (unusual sign-in, token replay) and executive account hardening matter most.
- Week 3, BEC: from the finance director's mailbox, the attacker creates a rule hiding replies containing "bank", then replies inside a genuine supplier thread with "updated remittance details". Control check: the inbox rule alert, the Reply-To mismatch, the supplier master change log and, decisively, the call-back to the supplier's number on file. If the payment has gone, the response clock starts: contact the bank and file with IC3 or Action Fraud at once, then follow the Business Email Compromise Response Playbook.
The lesson: had the organisation invested only in an email gateway, stages 2 and 3 would have been invisible to it. Had it invested only in finance process, stage 3 would have been stopped but the attacker would still hold two mailboxes.
Deepfakes and multi-channel pretexts
Whaling and BEC increasingly leave email. In the Arup deepfake fraud, a finance worker who first suspected a phishing message was persuaded by a video call in which the chief financial officer and other colleagues were deepfakes, and made 15 transfers totalling HK$200 million. The controls that would have helped were process controls, not email controls: call-back verification to a number on file, a second approver for unusual payments, and a rule that a recognised face or voice is not authentication. ATT&CK now models voice-based delivery as T1566.004 Spearphishing Voice. See the Executive Deepfake and CEO Fraud Guide.
Decision guidance: where to invest first
- If initial access incidents dominate: lead with phishing-resistant MFA for privileged and high-exposure staff, attachment and link controls, and blocking of illegitimate OAuth consent. This addresses spear phishing and the account takeovers that feed whaling and BEC.
- If you move large or frequent payments: lead with payment process controls. They are cheap, they work regardless of how good the forgery is, and they are the only control that stops a request sent from a genuinely compromised supplier.
- If your executives are publicly visible: harden executive and assistant accounts first (phishing-resistant MFA, mailbox auditing, delegate reviews), add display-name impersonation rules for each executive name, and run a digital footprint review.
- Small organisations: start with DMARC enforcement, MFA on every mailbox, and a written call-back rule for any bank detail change. These three cost little and cover the most common losses.
- Large organisations: add a supplier master change workflow with segregation of duties, automated mailbox rule detection in the SOC, lookalike domain monitoring, and role-based simulations so finance staff are tested on BEC scenarios rather than generic link clicks.
Measuring whether it works
- DMARC enforcement coverage: percentage of owned sending and non-sending domains at p=reject.
- Phishing-resistant MFA coverage: percentage of executives, assistants and finance users enrolled, with legacy protocols blocked.
- Payment verification compliance: percentage of bank detail changes with a documented call-back to a number on file, sampled monthly.
- Report rate by role: share of finance and executive staff who report role-specific simulations, not just overall click rate.
- Time to contain mailbox compromise: from first suspicious rule or sign-in alert to session revocation and rule removal.
- Time to bank contact after a fraudulent payment: measured from discovery, because recovery depends on it.
Framework mapping
| Framework | Reference | Relevance |
|---|---|---|
| MITRE ATT&CK | T1566 and sub-techniques, T1598, T1534, T1684.001, T1586.002, T1114.003, T1564.008, T1657 | Delivery, impersonation, mailbox persistence and financial theft |
| MITRE ATT&CK mitigations | M1017 User Training, M1054 Software Configuration | Role-based training; email authentication and client configuration |
| NIST CSF 2.0 | PR.AT-01, PR.AT-02 | General awareness, and specialised training for finance and executive roles |
| NIST CSF 2.0 | DE.CM-09 | Monitoring of mailbox and sign-in activity for adverse events |
| NIST SP 800-53 Rev. 5 | AT-2(3) Social Engineering and Mining, SI-8 Spam Protection, IA-2 Identification and Authentication, IR-4 Incident Handling | Training, mail filtering, MFA for users, response |
| ISO/IEC 27001:2022 Annex A | A.6.3 Information security awareness, education and training; A.5.14 Information transfer | Awareness programme and rules for transferring information, including payment instructions |
Common failure patterns
- One awareness programme for everyone: generic link-click simulations teach finance staff nothing about bank detail change fraud, and executives are often exempted from simulations entirely.
- Treating DMARC as BEC protection: an enforcing policy on your own domain does nothing when the request comes from a lookalike domain or a supplier's real, compromised mailbox.
- Verification inside the attacker's channel: replying to the email or calling the number in the email signature to "confirm" simply reaches the attacker.
- Executive exceptions: senior leaders exempted from MFA prompts, mailbox auditing or payment rules become the most valuable accounts to compromise.
- Trusting authentication results: a message that passes SPF, DKIM and DMARC proves only which domain sent it, not that the person behind it is genuine, as the TruffleNet case showed.
- Slow escalation after payment: waiting days to tell the bank wastes the window in which funds can be frozen.
Practitioner checklist
- Classify recent incidents as spear phishing, whaling or BEC and check that each type has an owning control.
- Move every owned domain, including parked ones, to DMARC p=reject with np=reject for non-existent subdomains.
- Enforce phishing-resistant MFA for executives, their assistants and all finance and payroll users first.
- Add display-name and lookalike domain impersonation rules for each executive and top suppliers.
- Alert on new inbox rules, external forwarding and delegate changes, and test the alert quarterly.
- Write a payment verification rule that no seniority can waive, with call-back to numbers on file.
- Run role-specific simulations: link and consent lures for all staff, pretext lures for executives, bank detail change scenarios for finance.
- Pre-agree the bank contact route and IC3 or Action Fraud reporting steps so a fraudulent payment is reported within hours.
Standards and references
- FBI IC3: 2024 Internet Crime Report
- FBI IC3: Public Service Announcement on Business Email Compromise (PSA240911)
- NCSC: Whaling: how it works, and what your organisation can do about it
- IETF: RFC 9989 Domain-based Message Authentication, Reporting, and Conformance (DMARC)
- MITRE ATT&CK: T1566 Phishing
- MITRE ATT&CK: T1684.001 Impersonation
- MITRE ATT&CK: T1657 Financial Theft
- NIST: SP 800-53 Rev. 5
Related NHI Mgmt Group resources: Phishing Guide · Email Identity and BEC Guide · Executive Deepfake and CEO Fraud Guide · Business Email Compromise Response Playbook