Substack data breach: why email-and-phone leaks matter (and what to do next)

When a company says “only emails and phone numbers were exposed,” it’s easy to shrug. No passwords, no credit cards — so what’s the harm?

In reality, email addresses and phone numbers are two of the most powerful join keys in the modern internet. They’re the identifiers that let attackers connect your presence across services, target you with convincing phishing and smishing, and in some cases attempt account takeovers through password resets or SIM swap–style social engineering. A leak of contact info isn’t the same as a leak of bank details, but it can be the first domino that makes other attacks cheaper and more likely to succeed.

This week, Substack began notifying some account holders that a security incident allowed an unauthorized party to access limited user data — including email addresses and phone numbers — with Substack saying the access occurred in October 2025 and was discovered and investigated in early February 2026. Substack’s CEO apologized and said that passwords and financial information were not accessed.

Let’s unpack what “limited user data” can still enable, what users should do (in order of impact), and what a good platform response looks like when the leaked fields are “just” contact details.

What Substack says happened (and what’s still unknown)

According to Substack’s notification message shared publicly and reported by The Verge, the company identified evidence of an issue that allowed an unauthorized third party to access limited user data without permission, including email addresses, phone numbers, and internal metadata. Substack says it fixed the problem and is conducting a full investigation, and that credit card numbers, passwords, and financial information were not accessed.

Two parts of that statement matter for understanding risk:

Timing: access occurred months earlier (October 2025) but was identified later. That gap can affect how long an attacker had to explore systems or exfiltrate data.
Scope uncertainty: the company did not publicly specify how many users were affected, what “internal metadata” includes, or whether any account state (like login history, device fingerprints, or subscriber lists) was exposed.

Those unknowns don’t automatically mean the worst. But they do mean users should assume that their email and phone number may be in the hands of someone who is willing to exploit them — and that their online accounts that use those identifiers may see an uptick in targeted scams.

Why email + phone is a high-value leak even without passwords

Email addresses and phone numbers are special because they’re used for three critical jobs:

Identity: they’re the handle many services use as your username.
Recovery: they’re the channels used to reset passwords.
Reach: they’re how scammers can contact you cheaply at scale (email) or with high urgency and credibility (SMS/phone).

Once an attacker has your email and phone, they can do more than send generic spam. They can tailor messages that reference the breached service, create pressure (“your account will be suspended”), and funnel you into a fake login page or a fake “verify your phone” flow. The goal isn’t necessarily to break into Substack; it’s to break into whatever else you use that shares the same contact info.

This is why agencies like CISA emphasize skepticism toward unsolicited messages, especially those that attempt to create urgency, request personal information, or push you to click a link. Phishing and its SMS variant (“smishing”) are often the cheapest way for an attacker to turn leaked contact info into money or access.

The practical risks for Substack users

Here are the most likely follow-on threats after a contact-info leak, and how they show up in real life.

1) Phishing emails that look like Substack

Expect emails that claim you need to “confirm your password,” “re-verify your payout details,” or “avoid losing access.” They may include official-looking logos and plausible sender names. The trick is usually in the domain (a lookalike URL) or in a link that goes somewhere other than substack.com.

What makes this worse: if attackers also have “internal metadata,” they may know whether you’re a writer vs. a reader, or what newsletters you subscribe to, and can make messages more convincing.

2) Smishing and vishing (texts and calls)

Leaked phone numbers invite a different style of scam: short, urgent texts and calls that exploit the fact that people tend to trust phone communications more than email. CISA notes that attackers can combine voice and text (“vishing” and “smishing”) with spoofed caller ID, pushing victims to disclose sensitive information or install “support” software.

A typical smish might say: “Substack Security: unusual login detected. Reply YES to lock your account.” If you reply, you’ve confirmed the number is active — and you’re engaged.

3) Account takeover attempts via password reset elsewhere

If you reuse passwords (or use weak ones), contact info becomes a way in. Even if your password wasn’t leaked from Substack, an attacker can attempt password resets on other sites that use your email or phone number. They don’t need the old password if they can convince you to hand over a code, or if they can compromise your email account.

This is why the “real crown jewels” are usually your email inbox and your mobile number: control either of those, and an attacker can often pivot into many other accounts.

4) SIM swap / number-port fraud attempts

SIM swapping isn’t new, but a dataset that ties phone numbers to a specific online service can help attackers pick targets. Many mobile carriers have improved defenses, but social engineering remains a risk, especially if an attacker can gather additional context from other sources.

If you rely on SMS for two-factor authentication (2FA) on important accounts, a SIM swap is one of the scariest outcomes. That’s not a reason to panic — it’s a reason to reduce dependence on SMS for critical logins where you have better options.

What you should do now (highest impact first)

Substack’s notification encourages extra caution around suspicious emails or texts. That’s good advice, but it’s vague. Here’s a more concrete, prioritized checklist that most people can execute in 15–30 minutes.

1) Lock down your primary email account

Your email inbox is the master key for password resets. If you do only one thing, do this:

Turn on two-factor authentication for your email provider (prefer an authenticator app or hardware key).
Review account recovery options (backup email/phone) and remove anything you don’t control.
Change your email password if it’s reused anywhere else.

If you’re a writer who uses Substack professionally, treat your email security like you treat your bank: it’s infrastructure.

2) Change your Substack password (and stop reusing passwords)

Even if Substack says passwords weren’t accessed, changing your password is cheap insurance — especially if you’ve used the same password on other sites. Use a unique, long password generated by a password manager.

If Substack supports passkeys or stronger login options, consider enabling them. The long-term goal is to make phishing less effective by reducing reliance on shared secrets you can type into a fake website.

3) Enable stronger 2FA where it matters most

On your most important accounts (email, banking, cloud storage, social media handles that can be used for scams), prefer:

Authenticator app (TOTP) over SMS
Hardware security keys (best, when available)
Passkeys (increasingly strong against phishing, depending on implementation)

SMS-based 2FA is better than nothing, but it’s vulnerable to number-port attacks and targeted social engineering. If you’re a public-facing creator, it’s worth upgrading.

4) Adjust your mental spam filter for a few weeks

After a breach disclosure, attackers often strike while the story is fresh. For the next month or so:

Be skeptical of messages that mention Substack, newsletters, “verification,” or “security.”
Don’t click login links from emails/texts. Instead, type the domain yourself or use a bookmark.
Be wary of urgent requests for one-time codes. No legitimate company support should ask for your 2FA codes.

CISA’s guidance is blunt for a reason: the easiest way to avoid being phished is to assume any unsolicited request could be hostile until you independently verify it.

5) If you’re a Substack writer: consider your audience-facing risk

Creators have an extra threat: impersonation. If attackers can convincingly spoof your Substack-related identity, they may target your subscribers with “special offers,” crypto scams, or fake fundraising links.

Consider posting a short note to your subscribers: you will never ask for passwords, one-time codes, or payment details over email.

What a good company response looks like (beyond an apology)

Substack’s message includes a fix and an investigation. That’s necessary, but the best breach responses are also specific. The FTC’s guidance for businesses emphasizes securing operations, fixing vulnerabilities, and communicating clearly without misleading omissions. In consumer terms, “clear” means:

Exactly what data fields were accessed (email, phone, name, address, IP logs, subscriber lists, DMs, etc.)
How many accounts were affected, even as a range
What users can do that meaningfully reduces risk (2FA, password manager, watch for smishing)
How the incident happened at a high level (e.g., exposed credential, vulnerable endpoint, misconfiguration) once it’s safe to disclose
What indicators of compromise exist (were there suspicious logins? can users view recent sessions?)

Companies sometimes avoid details because they don’t want to aid copycat attackers or because investigations are ongoing. But withholding too much leaves users guessing — and guessing tends to produce either panic or complacency.

Why “internal metadata” matters more than it sounds

In many breaches, the most damaging information isn’t the obvious fields. “Internal metadata” can be things like:

Account creation date and last login time
Whether an email is verified
Newsletter ownership or subscriber counts
Phone verification status
Support ticket history

Each of these helps an attacker craft a message that feels real. If a text says “Your Substack account created in 2021 needs re-verification,” and that’s true, the target is more likely to comply. That’s why transparency about what metadata was exposed is not a nerdy detail — it’s directly tied to how convincing the next wave of scams will be.

What you should not do

Three common mistakes after a breach disclosure:

Don’t reply to random “security” texts. You’re validating your number and inviting more contact.
Don’t install “support” apps because someone called you. If you need help, initiate contact through official channels you look up yourself.
Don’t assume you’re safe because you didn’t get an email. Notifications can be staged, incomplete, or limited by contact details on file. Treat this as a general warning sign, not a personal all-clear.

The bigger lesson: contact data is now “sensitive” data

We tend to rank data by whether it can directly drain a bank account. But the modern scam economy often works indirectly: leak identifiers, phish credentials, take over an email, reset other accounts, monetize access. Email addresses and phone numbers are the first rung of that ladder.

For platforms, this means that protecting contact data requires the same seriousness as protecting passwords — not because the fields are equivalent, but because attackers can turn them into leverage.

For users, it’s a reminder that the best defense isn’t perfect secrecy (breaches happen), it’s resilience: unique passwords, strong 2FA, and habits that make phishing unprofitable.

Bottom line

A leak of emails and phone numbers may sound “limited,” but it can still meaningfully increase your exposure to targeted phishing, smishing, and account takeover attempts elsewhere. Treat Substack’s notice as a prompt to harden the two systems that matter most: your email account and your login hygiene.

And if you see a sudden spike in “Substack security” messages, assume the incident has created a new scam wave — and don’t click first, verify first.