# Spear phishing vs phishing: targeted attacks

> What spear phishing is, how it differs from mass phishing, why AI makes a tailored message cheap to write, the signs, and how to stop it.

Source: https://trotta.email/threats/spear-phishing

Spear phishing is phishing written for one target. The attacker uses what is known about the target, such as a role, colleagues, suppliers and recent events, to write a message that reads as expected mail. CISA describes spearphishing as phishing aimed at an individual that includes key information about them.

Updated November 20, 2026

## Spear phishing vs phishing

Spear phishing differs from ordinary phishing in aim and effort. Ordinary phishing sends one message to many targets. Spear phishing sends one researched message to one target.

| What differs | Phishing | Spear phishing |
| --- | --- | --- |
| Targets | Many, chosen at random | One, chosen on purpose |
| Content | Generic: a parcel, a password, a prize | Specific: real names, projects and dates |
| Sender | A well-known brand | A colleague, supplier or client |
| Goal | Passwords in bulk | One account, one payment or one file |

## How spear phishing works

Spear phishing works by research first and writing second.

1. **Choose the target.** The attacker picks an employee with access: finance, HR, IT or an executive's assistant.
2. **Collect the details.** Public profiles and company pages give names, roles, suppliers and travel.
3. **Pick a pretext.** The message follows a real event, such as a meeting, an order or a job posting.
4. **Pose as a known sender.** The attacker uses a [lookalike domain](https://trotta.email/threats/lookalike-domains) or a taken-over mailbox.
5. **Ask for one thing.** A sign-in, a file to open or a payment. [Credential phishing](https://trotta.email/threats/credential-phishing) and [CEO fraud](https://trotta.email/threats/ceo-fraud) often start this way.

## What spear phishing looks like

A spear phishing message reads like mail the target was waiting for. In this fictional example, the sender poses as Megan Park of Northwind, a supplier Jordan Ellis met last week.

Example only, not a real message:

```text
From: Megan Park <megan@northwlnd-supply.example>
To: Jordan Ellis <jordan@trotta.email>
Subject: Price list from Thursday

Jordan, good to meet you at the supplier review on Thursday.
The price list we discussed is at the link below.
Can you check it before your Monday call?
[link to a page that asks for a sign-in]
```

The meeting, the day and the call are real. The sender's domain and the link are not.

## Signs of spear phishing

| Sign | What it means |
| --- | --- |
| Correct personal details from a new address | Public details do not prove who is writing. |
| A known name on a near-match domain | The attacker registered a lookalike. |
| A link or file that needs a sign-in to view | The pretext exists to collect a password. |
| A request that fits the job but skips the usual route | The attacker studied the process. |
| Timing right after a meeting, a post or a trip | The message was prompted by something public. |

## Why AI makes spear phishing cheaper

AI makes spear phishing cheaper because the costly part, research and writing, is now quick. A language model can read a target's public profile and draft a fluent, personal message.

Work that once limited an attacker to a few targets can now cover a whole staff list. The FBI IC3 2025 report notes that chat tools can quickly write official-sounding emails that mimic a company's executives. CISA notes that in the era of AI some phishing emails have perfect grammar.

## Why spear phishing gets through ordinary filtering

Spear phishing gets through ordinary filtering because each message is one of a kind. Filters learn from volume, and a message sent once has no pattern to match.

The lookalike domain passes [SPF, DKIM and DMARC](https://trotta.email/learn/spf-dkim-dmarc), and the link is fresh. Gmail and Outlook screen mail before delivery, yet a single message that looks like real work passes. The Verizon 2026 Data Breach Investigations Report found the human element involved in 62% of breaches.

## When an AI agent reads the mail

An AI agent can be spear phished too, and what an attacker needs to know about it is easy to find. An agent's address and its job are often public, as with a [support agent](https://trotta.email/use-cases/customer-support).

A message tailored to an agent does not need to deceive a human eye. The message can address the model directly, in the open or in hidden text. See [prompt injection by email](https://trotta.email/threats/prompt-injection-by-email) and [AI agent security](https://trotta.email/learn/ai-agent-security).

## How to stop spear phishing

Spear phishing is stopped by checks that hold even when the message is flawless.

- **Verify by a second channel.** Confirm an unusual request by phone or chat, using details already on file.
- **Use phishing-resistant sign-in.** A stolen password should not be enough.
- **Limit what is public.** Publish less about who approves payments and which suppliers you use.
- **Require approvals.** Payments and data exports need a second human.
- **Train with tailored examples.** Show the finance team mail that uses their own names and suppliers.

### At the door, before delivery

A pre-delivery inbox takes each message apart before anyone reads it. Trotta Mail checks the sender (SPF, DKIM and DMARC, lookalike domains, first contact), the contents (where links go and what files are) and text aimed at a model.

Dangerous mail is [held](https://trotta.email/learn/held-mail) for a human, and mail worth a second look is delivered with the caution attached. The guard is stricter for an agent than for a human, and an agent's API key cannot read or release held mail. See the [security model](https://trotta.email/docs/security).

No filter catches everything. A tailored message with no technical tell can still be delivered, which is why uncertain mail is held or flagged for a human.

## Questions

### What is the difference between phishing and spear phishing?

Phishing sends the same generic message to many targets. Spear phishing sends a researched message to one target, using real names, roles and events.

### What is whaling?

Whaling is spear phishing aimed at a senior executive. CISA describes whaling as phishing aimed at a high-profile individual to steal sensitive, high-value information.

### How do attackers find information for spear phishing?

Mostly from public sources: professional profiles, company pages, press releases and social posts.

### How does AI change spear phishing?

AI lowers the cost of each tailored message, so an attacker can send researched mail to many employees. Errors in grammar are no longer a dependable sign.

## Sources

- [CISA, Phishing: general security postcard](https://www.cisa.gov/sites/default/files/publications/Phishing%20General%20Security%20Postcard_6.24.2021_508cV2.pdf)
- [CISA, Recognize and Report Phishing](https://www.cisa.gov/secure-our-world/recognize-and-report-phishing)
- [FBI Internet Crime Complaint Center, 2025 Internet Crime Report](https://www.ic3.gov/AnnualReport/Reports/2025_IC3Report.pdf)
- [Verizon 2026 Data Breach Investigations Report, executive summary](https://www.verizon.com/business/resources/executivebriefs/2026-dbir-executive-summary.pdf)
- [Trotta Mail docs: Concepts](https://trotta.email/docs/concepts)

## Related

- [Credential phishing](https://trotta.email/threats/credential-phishing) Threat library
- [Business email compromise (BEC)](https://trotta.email/threats/business-email-compromise) Threat library
- [Prompt injection by email](https://trotta.email/threats/prompt-injection-by-email) Threat library
- [Customer support](https://trotta.email/use-cases/customer-support) Use cases
- [AI agent security](https://trotta.email/learn/ai-agent-security) Guides
- [Security model](https://trotta.email/docs/security) Docs

Every check is on, on every plan. Free for two inboxes. No card.

[Get an inbox](https://trotta.email/signup) [Read the security model](https://trotta.email/docs/security)
