Skip to content
Vol. 1 · Ed. 2026
CyberGlossary
Entry № 580

Impossible Travel Detection

Reviewed byCybersecurity entrepreneur & security researcher

What is Impossible Travel Detection?

Impossible Travel DetectionA detection that flags successive sign-ins from geographic locations too far apart to be reached by any plausible travel within the elapsed time.


Impossible travel detection compares the IP-geolocated source of two successful authentications for the same identity and computes the implied travel speed. If user U signs in from New York at 09:00 UTC and from Singapore at 10:30 UTC, the implied velocity far exceeds commercial aviation, so the second event is scored as risky. The analytic is a cheap, high-signal way to surface account takeover and stolen-session reuse, and it maps to MITRE ATT&CK technique T1078 (Valid Accounts).

Microsoft Entra ID (Identity Protection), Okta ThreatInsight, Google Workspace and most SIEM/UEBA platforms ship a built-in "atypical travel" or "unusual location" rule. Mature implementations do not rely on raw distance alone: they learn each user's familiar locations and devices, and suppress known false-positive sources — corporate VPN egress points, cloud/mobile-carrier NAT, and Tor exit nodes — because a VPN that lands traffic in another country is the single biggest cause of noise.

Because the signal is probabilistic, playbooks treat it as a trigger, not a verdict: raise the session's risk score, force step-up authentication or a password/MFA re-prompt, or revoke tokens and sessions. Correlating impossible travel with other weak signals (new device, MFA-fatigue prompts, mailbox-rule creation) sharply reduces false positives while catching real intrusions early.

flowchart TD
  A["Sign-in event 1<br/>New York, 09:00 UTC"] --> C["Compute distance / elapsed time"]
  B["Sign-in event 2<br/>Singapore, 10:30 UTC"] --> C
  C --> D{"Implied speed > threshold<br/>(e.g. 800 km/h)?"}
  D -->|"No"| E["Normal — allow"]
  D -->|"Yes"| F{"Known VPN / Tor / carrier NAT?"}
  F -->|"Yes"| G["Suppress — likely false positive"]
  F -->|"No"| H["Risk score up → step-up MFA<br/>or revoke session/token (T1078)"]

● Examples

  1. 01

    Entra ID risky sign-in raising 'atypical travel' when the same account logs in from Paris and Sao Paulo within 30 minutes.

  2. 02

    An Okta ThreatInsight policy revoking sessions when geo-velocity exceeds 800 km/h.

● Frequently asked questions

What is Impossible Travel Detection?

A detection that flags successive sign-ins from geographic locations too far apart to be reached by any plausible travel within the elapsed time. It belongs to the Identity & Access category of cybersecurity.

What does Impossible Travel Detection mean?

A detection that flags successive sign-ins from geographic locations too far apart to be reached by any plausible travel within the elapsed time.

How do you defend against Impossible Travel Detection?

Defences for Impossible Travel Detection typically combine technical controls and operational practices, as detailed in the full definition above.

What are other names for Impossible Travel Detection?

Common alternative names include: Atypical travel, Geo-velocity anomaly, Superman travel.

● Related terms