Spear phishing is a targeted form of phishing in which an attacker researches a specific person or small group and crafts a personalized deceptive message to make the victim open a malicious file, follow a harmful link, reveal credentials or authorize an action. In cybersecurity, it belongs to the malware and attacks family as one of the most common initial access techniques in intrusions attributed to both criminal groups and state-sponsored actors.
Unlike bulk phishing, which relies on volume and expects a tiny response rate, spear phishing relies on credibility built from reconnaissance. A single well-crafted email to the right employee can deliver ransomware, an infostealer or a remote access trojan into a network that technical controls otherwise protect well.
How a spear phishing attack unfolds
A spear phishing operation follows a repeatable attack chain that starts long before any message is sent. The attacker invests most of the effort in reconnaissance and pretext design, because the quality of the lure determines success.
| Stage | Attacker activity | Typical artifacts |
|---|---|---|
| Target selection | Identifies people with useful access or authority | Org charts, job postings, conference speaker lists |
| Reconnaissance | Collects names, projects, contacts and writing style | Professional networking profiles, press releases, leaked data |
| Pretext design | Builds a believable reason to act | Fake invoice, shared document, recruiter offer |
| Infrastructure setup | Registers lookalike domains and hosting | Typosquatted domains, cloned login pages |
| Delivery | Sends the lure through email, chat or social media | Weaponized attachment or credential link |
| Exploitation | Runs code or harvests credentials after the click | Macro loader, session token theft |
The final stage usually leads to follow-on activity: installing persistent malware, moving laterally, or logging into cloud services with stolen credentials. Spear phishing is therefore a door-opener rather than the end goal of the intrusion.
Variants and delivery techniques
Spear phishing has several recognized variants that differ by target, channel or payload. Security teams use these labels to choose matching controls and training content.
| Variant | Target or channel | Distinctive feature |
|---|---|---|
| Whaling | Senior executives and board members | Lures reference legal, financial or strategic matters |
| Clone phishing | Recipients of a genuine earlier message | Resends a real email with a malicious replacement link |
| Thread hijacking | Participants in an existing conversation | Replies inside a compromised mailbox thread |
| Spear vishing | Phone calls to specific staff | Voice pretext, sometimes with cloned voices |
| Smishing and chat lures | Mobile messaging and collaboration tools | Bypasses email filtering entirely |
| Callback phishing | Employees who handle subscriptions or billing | Email contains no link, only a phone number |
Attackers increasingly combine channels, for example an email followed by a phone call that confirms the pretext. Generative AI tools lower the cost of fluent, context-aware text, so grammar mistakes are no longer a reliable warning sign.
Spear phishing compared with related attacks
Spear phishing overlaps with several neighboring concepts, and the differences lie in targeting, payload and objective. Precise terminology matters because each attack calls for different controls.
| Attack | Targeting | Primary objective |
|---|---|---|
| Spear phishing | Specific individuals, researched | Initial access through malware or credentials |
| Bulk phishing | Thousands of generic recipients | Credential harvesting at scale |
| Business email compromise | Finance staff and executives | Fraudulent payments, often without malware |
| Social engineering | Any person, any channel | Umbrella term for psychological manipulation |
| Watering hole attack | Visitors of a site the group trusts | Malware delivery without direct messaging |
Business email compromise frequently begins with spear phishing that steals a mailbox password, then continues as a fraud scheme. In threat frameworks such as MITRE ATT&CK, spear phishing appears under the Phishing technique with sub-techniques for attachments, links and third-party services.
Detecting and defending against targeted lures
No single control stops spear phishing, so effective defense relies on layered technical and human measures. The goal is to make delivery harder, make the click less harmful and make reporting fast.
| Control | What it stops or limits | Known limitation |
|---|---|---|
| SPF, DKIM and DMARC enforcement | Exact spoofing of the organization’s own domain | Does not block lookalike domains |
| Phishing-resistant MFA | Use of stolen passwords and relayed codes | Requires FIDO2 keys or passkeys, not SMS codes |
| Attachment sandboxing and link rewriting | Known malicious payloads and URLs | Evaded by delayed or conditional payloads |
| Endpoint detection and response | Malware execution after a click | Depends on tuning and analyst response |
| One-click reporting button | Dwell time of campaigns in mailboxes | Works only if employees trust the process |
| Out-of-band verification rules | Fraudulent payment or access requests | Must be followed even under time pressure |
Security awareness training helps most when it teaches staff to check the request rather than the cosmetics of the message. Simulated campaigns are useful for measuring report rates, which matter more than click rates, because one early report can let the security team purge the same lure from every mailbox.
Misconceptions and organizational blind spots
Organizations often underestimate spear phishing because of persistent misconceptions about who is targeted and what a lure looks like:
- Only executives are targets. Assistants, help desk agents and recruiters are frequent targets because they open external messages as part of their job.
- Suspicious messages look suspicious. Well-researched lures reference real projects, real colleagues and correct internal jargon.
- MFA solves the problem. Adversary-in-the-middle phishing kits relay one-time codes in real time and steal session cookies.
- Blaming the clicker improves security. Punishment reduces reporting, which slows incident response and hides campaigns.
The strongest programs treat a click as an expected event to contain, not a personal failure, and design systems so that one compromised user cannot reach critical assets alone.
The idea in brief
Ordinary phishing is like a fisherman casting a wide net and hoping something swims in. Spear phishing is aiming at one particular fish. The attacker first learns about you: your job, your manager’s name, the project you are working on, perhaps a recent trip. Then they send a message that seems to come from someone you know, about something you expect.
Because the message feels familiar, it is much easier to trust. The best protection is a simple habit: when a message asks you to open something unexpected, log in, or move money, confirm it through a separate channel you already know, such as a phone number from the company directory.
A practical scenario
An engineering firm announces on its website that it has won a bridge design contract. Within days, an attacker studies the project team on professional networking sites and registers a domain that differs from the client’s real domain by one swapped letter. A project engineer then receives an email, apparently from the client’s procurement officer, with a file named as the revised structural requirements.
The attachment is a disk image containing a shortcut file that launches an infostealer. The engineer opens it, but the endpoint detection tool flags the unusual script activity within minutes and isolates the laptop. Because the engineer also pressed the report button, the security team finds the same email in four other mailboxes and removes it before anyone else opens it.
The firm resets the engineer’s credentials, revokes active sessions, and adds the lookalike domain to its block list. The outcome is a contained incident with one isolated device and no stolen design data.