Back to articles

Account takeover prevention: detecting credential theft campaigns

The short answer

Account takeover prevention has two halves. Inside your perimeter, phishing-resistant MFA, conditional access, and session controls make stolen credentials harder to use. Outside it, you find and remove the campaigns that collect those credentials: fake login pages, adversary-in-the-middle proxies, poisoned search ads, fake support accounts, and leaked credentials for sale. Most ATO programs cover the first half and miss the second.

Key takeaways
  • ↗The FBI received more than 5,100 ATO complaints with over $262 million in losses from January to November 2025, from criminals impersonating financial institution support.
  • ↗Adversary-in-the-middle phishing steals session cookies, so standard MFA alone does not stop it.
  • ↗Credential theft campaigns leave signals outside your network: lookalike domains, brand-bidding ads, fake support profiles, and infostealer logs.
  • ↗Removing the harvesting infrastructure early shortens the window in which stolen credentials can be collected.

Why account takeover now starts outside your network

In November 2025 the FBI warned that criminals were impersonating financial institution support staff to take over customer accounts. It counted more than 5,100 complaints and over $262 million in losses since January 2025. The methods it lists include social engineering for login credentials and one-time codes, phishing websites that look like real banking portals, and search engine ads that push those sites above the real ones.

Workforce accounts face the same pattern. In Verizon's 2025 Data Breach Investigations Report, 30% of systems in infostealer logs were enterprise-licensed devices, and 46% of systems holding corporate logins were unmanaged. Among ransomware victims, 54% had their domains appear in credential dumps.

The login is where the takeover becomes visible. The theft happened earlier, on a fake page, a malicious ad, or an infected personal laptop your security tools never saw.

The credential theft campaign, stage by stage

01

Lure

Search ads on brand and login terms, SMS and email phishing, and fake support accounts on social platforms send victims to the attacker.

02

Harvest

A cloned login portal collects usernames and passwords. An adversary-in-the-middle proxy relays the real sign-in and captures the session cookie, so the attacker skips MFA.

03

Package and resell

Infostealer malware and phishing kits feed logs and credential lists that are sold or shared on markets and messaging channels.

04

Take over

The attacker replays the session or signs in, then adds an MFA method, changes recovery details, or resets the password to lock the owner out.

05

Monetize

Funds are moved, gift cards or loyalty points are drained, or the account is used for business email compromise against contacts.

Microsoft documented one adversary-in-the-middle campaign that targeted more than 10,000 organizations from September 2021 and used stolen sessions for follow-on payment fraud. In a later multi-stage campaign, attackers added a new MFA method after replaying a stolen session, then sent more than 16,000 phishing emails from the compromised account. See Microsoft's analysis.

Signals to watch outside the perimeter

SignalWhere it shows upWhat it tells you
Lookalike login domainsNew registrations and certificate transparency logsA harvesting page is being staged or is live
Mail records on lookalikesDNS MX recordsThe domain can send phishing that appears to come from you
Ads on brand and login termsSearch and social ad librariesTraffic is being bought to send victims to a fake portal
Fake support profilesSocial platforms, app reviews, messaging appsVictims who ask for help are steered to attackers
Kits that copy your assetsPhishing pages loading your logos, CSS, or scriptsA reusable kit is being deployed across many domains
Your domains in stealer logsDark web markets, combolists, messaging channelsCredentials and sessions for your users are already out

Controls inside the perimeter

External detection does not replace identity security. These controls make the credentials that do get stolen harder to use.

01Use phishing-resistant MFA such as FIDO2 security keys or passkeys for high-risk users
02Apply conditional access based on device compliance, location, and risk
03Shorten session lifetimes and revoke sessions after password resets
04Alert on new MFA methods, recovery changes, and inbox rules
05Reset accounts found in infostealer logs or credential dumps
06Tell customers which channels you use for support, and which you never use

Response playbook when you find a campaign

01

Confirm and capture

Screenshot the page, record the URL, and save the kit's source and any form endpoints before reporting.

02

Remove the infrastructure

Report the domain to the registrar, the site to the host and CDN, and the ads and support profiles to each platform, in parallel.

03

Warn browsers

Submit URLs to browser blocklists so visitors see a warning while removal is pending.

04

Protect exposed accounts

Work with identity and fraud teams to reset credentials, revoke sessions, and watch affected accounts.

05

Watch for the next deployment

Link the campaign's hosts, name servers, and kit fingerprints, and monitor for the same pattern on new domains.

For the takedown steps in detail, see how to take down a phishing lookalike domain.

Where Outtake fits

Outtake platform

The external half of account takeover prevention

Outtake does not replace your identity provider or MFA. It finds and removes the infrastructure that collects credentials before they reach your login page, and links each finding to the wider campaign.

  • ↗Detects cloned login portals, fake support accounts, and brand-bidding ads
  • ↗Monitors dark web sources for leaked credentials and exposed data
  • ↗Recon Agent expands one phishing page into the campaign behind it
  • ↗Remediation Agents file takedowns and re-check until items are confirmed down

Learn more about the Outtake platform or the Recon Agent.

Frequently asked questions

What is account takeover?

Account takeover is when an attacker gains control of a legitimate user's account, usually with stolen credentials or a stolen session, then uses it for fraud, data theft, or further phishing.

Does MFA prevent account takeover?

MFA blocks many attacks, but adversary-in-the-middle phishing can steal the session cookie after the user completes MFA, and social engineering can trick users into sharing one-time codes. Phishing-resistant MFA such as FIDO2 or passkeys closes most of that gap.

How do I detect credential theft campaigns targeting my company?

Monitor outside your network: new lookalike domains and certificates, ads bidding on your brand and login terms, fake support profiles, phishing pages that load your assets, and your domains in infostealer logs or credential dumps.

What is an AiTM phishing attack?

In adversary-in-the-middle phishing, a proxy site sits between the victim and the real login page. It relays the sign-in, including MFA, and captures the session cookie, which the attacker replays to access the account.

Who should own account takeover prevention?

Identity and security engineering usually own authentication controls. Fraud, brand protection, or threat intelligence teams usually own external detection and takedown. The two should share findings so exposed accounts are reset quickly.

Related resources

Sources