At 2:14 in the afternoon, your office manager signs in to Microsoft 365 from her desk in Daytona Beach. At 2:31, someone signs in to the same account from another country, with the right password and an approved multi-factor prompt. To the system, both look like her.
That’s what an identity attack looks like. Nobody broke anything. They signed in, using a password that was phished, guessed or bought, and often a session stolen in the same phishing attempt so they don’t need the second factor again. From there they read email, set up forwarding rules, look for invoices to redirect, and send convincing messages to your customers from a real account.
Identity threat detection and response (ITDR) is the set of tools and people that notice that second sign-in, and act on it.
Why identity is now the main target
Most of what a business runs now sits behind a sign-in: email, files, accounting, the bank portal. One account opens a lot of doors, so attackers go after accounts. Verizon’s 2026 Data Breach Investigations Report found that exploited software flaws had overtaken stolen passwords as the most common way attackers get in (Verizon DBIR, 2026).
Multi-factor sign-in stops most password attacks, and every business should have it. But attackers have adapted: fake sign-in pages that pass the multi-factor prompt through to the real site and keep the resulting session, repeated prompts sent until someone taps “approve” to make them stop, and calls to the help desk asking for a reset. Detection is the layer that catches what gets past the front door.
What identity threat detection watches for
The tools build up a picture of how each person normally signs in, and flag what doesn’t fit. Typical signs include:
- Impossible travel. Sign-ins from two places too far apart for the time between them.
- Unfamiliar sign-ins. A new country, a new device, a new browser, or an address known to hide where the person really is.
- A reused session. A sign-in token turning up from a different place than the one it was issued to, a sign of a stolen session.
- Password spraying. A common password tried against many accounts, and one of them working.
- Changes after sign-in. A new forwarding rule to an outside address, rules that hide or delete messages, a new multi-factor method added to the account, or a sudden burst of outgoing email.
Any one of these can have an innocent explanation; the person may really be on vacation. The value is in combining them and acting quickly on the ones that add up.
One account opens a lot of doors, so attackers go after accounts.
What happens when it fires
The response is what makes it more than a report. Depending on the tools and how they’re set up, a high-risk sign-in or account can trigger:
- Signing the account out everywhere, which ends any stolen session, not just the attacker’s current one.
- Blocking the account or requiring a password change and a fresh multi-factor check before anyone can use it again.
- An alert to whoever looks after your IT, with the details of what was seen.
- Investigation: what the attacker opened, what rules they created, what they sent, and whether any other accounts were touched.
- Cleanup: removing their rules and added sign-in methods, warning anyone who received messages from the account, and calling the bank if a payment may have been redirected.
The first two happen in seconds and limit the damage. The last three need a person who knows your business.
What Microsoft 365 includes, and what it doesn’t
If you use Microsoft 365, some of this is built in. Microsoft’s identity service flags risky sign-ins and risky users. The full version, which shows the details and can automatically block or challenge risky sign-ins, is called Microsoft Entra ID Protection and needs a Microsoft Entra ID P2 license; plans such as Business Premium include P1, which gives only limited risk reports (Microsoft Learn).
So a business on Business Premium has a choice: add the P2 license, use a separate identity detection product, or rely on someone reviewing the limited reports by hand. Which is right depends on your size and risk, but it’s worth knowing that the automatic version isn’t switched on by default.
How it fits with the rest of your security
Identity detection is one layer among several, and it works best alongside the others:
- Multi-factor sign-in on every account, preferably with number matching or a passkey rather than a simple approve button.
- Rules about where and how people can sign in, such as only from company devices.
- A company password manager, so one leaked password doesn’t open every account.
- Endpoint detection and response on every computer, which covers attacks on the device rather than the account. Our post on automated threat detection and response explains how that side works.
- Training, especially on the fake sign-in page and the surprise multi-factor prompt.
Four questions for your IT provider
- If someone signed in to one of our accounts from another country with the right password, would anyone know, and how quickly?
- Would the account be blocked automatically, or would it wait for a person?
- Who gets the alert, and what do they do next?
- Do we have the licensing for the detection we think we have?
For a closer look at how account takeover plays out in Microsoft 365, see our post on Microsoft 365 account takeover.
How we protect accounts
Identity is one of the layers in our cybersecurity setup for Managed IT customers: multi-factor sign-in, access rules and a company password manager, with identity threat detection and response that watches for unusual sign-ins and locks the account down. If something does get through, we help you contain it, lock down affected accounts and devices, and work out what happened. To find out how your accounts are protected today, our free technology review checks it and gives you a written report. Call (386) 252-2292.