Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
Account takeover is the compromise of a legitimate user identity, after which an attacker operates as that user. Detection depends on behaviour, and response starts by revoking sessions, not by resetting a password.

What Is Account Takeover?
Account takeover is the point at which an attacker stops knocking and starts using the front door. They hold working credentials, or a stolen session cookie, and from then on the activity in your logs belongs to a legitimate identity. Nothing is technically broken, which is why this is one of the hardest compromises to spot and one of the easiest to under-scope.
Password reuse remains the cheapest route: attackers replay credentials from old dumps against your login page, which is what credential stuffing describes. Infostealer malware such as RedLine and Lumma has made that worse, because the logs it sells contain live session cookies alongside passwords, and a cookie sidesteps multi-factor authentication entirely.
Adversary-in-the-middle phishing kits, Evilginx2 and EvilProxy among them, do the same thing in real time: the victim authenticates against a proxy that forwards every field to the real identity provider and keeps the resulting token. Where push-based MFA is in use, attackers simply spam approval prompts until someone taps accept, the technique that gave an intruder access to Uber in September 2022. And where the helpdesk can reset a factor over the phone, social engineering beats all of it, as the attacks on MGM Resorts in September 2023 demonstrated.
invoice, payment, or bank into Deleted Items or the RSS Subscriptions folder.Most of these come from the same place, which is why authentication log analysis is the backbone of any takeover detection programme.
Revoke before you reset. A password change alone leaves existing refresh tokens valid, and in Microsoft Entra ID those can live for up to 90 days, so the attacker stays signed in while you congratulate yourself. Revoke sessions, then reset the password, then re-enrol the factors from a known-good device.
Next, hunt the persistence that outlives the password: registered authenticators, app passwords, inbox and transport rules, mailbox delegation and forwarding, OAuth grants, personal access tokens, and SSH keys added to any linked developer platform. Attackers plant two or three of these within minutes precisely because defenders stop at the password.
Only then scope the damage. In Microsoft 365, the unified audit log and MailItemsAccessed events tell you which messages were opened; Entra sign-in logs default to a 30-day window, so export them before they age out. Record what you find as you go, because the same timeline will answer the regulator’s questions later.
If personal data was accessible, the clock in Article 33 of the UK GDPR starts when you become aware of the breach, and you have 72 hours to notify the Information Commissioner’s Office unless the risk to individuals is unlikely. A single mailbox counts: the ICO has acted on exactly that scenario. Fraud attempted through the account should go to Action Fraud, financial services firms have their own notification obligations to the FCA, and operators in scope of the NIS Regulations report to their competent authority. Keeping these separate from the technical response is easier if someone other than the incident lead owns them.
Phishing-resistant factors are the only control that reliably breaks the proxy kits, because FIDO2 and certificate-based authentication bind the credential to the origin. Conditional access policies that require a compliant device, token protection where it is supported, and short session lifetimes for high-privilege roles all shorten the window. Seeding the directory with canary accounts gives you a high-confidence alert when someone starts enumerating identities.
Our investigation and incident response services cover identity compromise end to end, including the audit log export that most teams realise they need on day four.
Because modern identity platforms issue refresh tokens that survive a password change until they are explicitly revoked. In Microsoft Entra ID a refresh token can remain valid for up to 90 days, so an attacker who already holds one keeps minting new access tokens. Always revoke sessions first, then reset the password, then re-register the authentication methods from a device you trust.
Look for a cluster rather than a single indicator. A genuine traveller signs in from a consumer or carrier network, keeps the same device identifier, and does not register a new MFA method or create a mail rule within minutes of arriving. Attackers usually arrive from hosting or proxy infrastructure and immediately take actions that establish persistence or search the mailbox.
No. Notification is required within 72 hours only where the breach is likely to result in a risk to the rights and freedoms of individuals. A compromised mailbox containing client correspondence will usually meet that bar; an account with no access to personal data may not. Document the assessment either way, because the ICO expects to see the reasoning behind a decision not to report.
Identity provider sign-in and audit logs, mail access and rule-change events, VPN and remote access logs, and any SaaS admin audit trail. Default retention is often 30 days, which is shorter than the time it typically takes to discover the compromise, so forward them to your SIEM or an archive tier. Twelve months is a reasonable target for most organisations.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.