Most account takeovers do not begin with someone guessing your password. They begin with a password that was already published in a breach and reused on a second site. Checking takes about five minutes, costs nothing, and does not require installing anything.

First, what a leak actually means

A breach is a copy of a company's database getting out. Passwords in that database may be scrambled, but scrambling is not a one-way door: weak and reused passwords get recovered quickly, and the results are traded in bulk lists that are tried against other services.

That gives you two separate questions. Was your email address exposed, which is about identity and spam? And has your password appeared in breach data, which is about someone being able to log in as you? The second is the urgent one.

Step 1: check the account

Go to Have I Been Pwned and search your email address. The service collects breaches and shows which ones include your address, with dates and the kinds of data exposed.

Two things to keep in mind. A result is a record of the past, not a live alarm: nothing there tells you whether someone is using your account right now. And an empty result is not proof of safety, because plenty of breaches are never disclosed.

Step 2: check the password itself

Have I Been Pwned also runs Pwned Passwords, which tells you whether a specific password appears in its breach corpus. It is designed so the full password never leaves your device: the page hashes it, sends only a short prefix of the hash, and matches the remainder locally. That is why it is safe to type a real password there and unsafe to type one into a random "leak checker" that appeared in an advert.

If the answer is yes, treat that password as public information.

Step 3: use the checkup you already have

Google Password Manager, reachable at passwords.google.com, includes a checkup that flags saved passwords which are compromised, reused or weak, and links straight to the site to change them. Most browsers and password managers offer the same thing, usually powered by the same breach data.

The limitation is obvious: it only knows about passwords you have saved. Anything typed from memory or kept in a notebook is invisible to it.

Step 4: change passwords in the right order

Start with the email account. It is the reset route for everything else, so an attacker who holds it can work outward from there. Then banking and payment accounts, then shopping, then social.

Three rules make the new password worth the effort:

  • Make it long and unique to that one site. Three random words is the approach the NCSC recommends, because length beats complexity you cannot remember.
  • Do not take the old password and add a number or a season. Cracking tools are built for exactly that pattern.
  • Change it on the site itself, by typing the address, not through a link in an email that told you to.

Step 5: turn on a second factor

A second factor turns a stolen password into a dead end. An authenticator app or a passkey is stronger than a text message, which can be intercepted or SIM-swapped, but SMS is still far better than nothing.

Where a service offers passkeys, take it. A passkey removes the password from the login altogether, so there is nothing to leak in the next breach.

Step 6: sign out everything else

Most services list active sessions under a "your devices" or "security" heading. Sign out anything you do not recognise, then change the password again if anything looked wrong.

In your email account, also check recovery addresses, recovery phone numbers and forwarding rules. Attackers who get into a mailbox routinely add a forwarding rule and then sit quietly, reading password resets as they arrive.

If you reused the password

This is the case that matters most. Reuse turns one breach into many, because automated tools try the same email and password pair across hundreds of sites, a technique known as credential stuffing.

Assume every account that shared that password is compromised. Change them all, prioritise the ones tied to money or identity, and watch for password reset emails you did not request: they mean somebody else is trying.

What not to do

  • Do not pay anyone who emails to say they have your password. It is almost always old breach data being recycled as a threat.
  • Do not trust pop-ups that claim to scan your device for leaks. Close the tab.
  • Do not email a password to yourself, or keep it in a notes app that syncs to an account you have just been locked out of.
  • Do not assume a single change is enough if the same password protected five accounts.

The five-minute version

Search your email on Have I Been Pwned. Check the password itself on Pwned Passwords. Change the email password first, then anything financial. Turn on a second factor. Sign out other sessions. If the password was reused, keep going until every account that shared it has a new one.