AI agents investigate every alert, add context, and prepare cases for response, on a full-stack SecOps platform that captures and correlates all your security data.
Cloud-based SIEM platform providing visibility across all your data sources
Every alert investigated, with a verdict and the evidence behind it
Manage your external footprint, publicly available data, and Dark Web leaks
Log management and threat detection across any of your log sources
Get ultimate assurance that your environment is free from threat actors
Fully featured distrubuted NDR for on-premises and public cloud environments
A SecOps platform purpose-built for industrial and OT environments
Integrated endpoint agent providing deep visibility, response and forensics
Deploy honeypots and honeytokens and stay ahead of threat actors
24x7 security operations from our Australian SOC. AI investigates every alert, and experienced analysts validate findings and contain threats around the clock.
24x7 managed detection and response across endpoint, network, cloud, identity, and log sources
AI investigations and 24x7 MDR on the Microsoft Sentinel you already own
Eyes-on-glass SOC and continuous threat hunting
Expert hunters and AI agents find attacker activity your alerts missed
Quickly recover from cyber attacks and breaches
ThreatDefence is the only SecOps as a Service company providing broad coverage across your entire technology stack with evidence-based security.
We provide security teams with full-stack SecOps infrastructure – you get deep visibility from day one, gain control over your security data, and get access to a comprehensive set of ready-to-use tools, detections, workflows, playbooks and scenarios.
24x7 SOC, Continuous Monitoring, Threat Containment and Incident Response
Visibility and actionable insights for all IT assets
Leverage our Cyber Range to train defenders based on real-world scenarios and simulations
OT and industrial cyber security services for critical infrastructure
Monitoring, evidence and a 24×7 SOC in one platform
Compliance monitoring, evidence and board reporting
Privacy Act reform impacts, obligations, and practical security considerations
Monitoring and compliance support for critical infrastructure environments
A practical guide to securing critical infrastructure with continuous monitoring
A practical buyer’s guide for schools evaluating SIEM and SOC services
A practical guide to selecting the right SIEM and SOC service for local government
Articles on security operations, AI and compliance
Events, awards and company news
Read what our customers say about us
Solution briefs and datasheets
How we keep your data secure
Technical support and knowledge base
Guides, case studies, and advisories in one place
How AI agents investigate alerts and run security operations
Free ebook: a practical guide to adopting an AI SOC platform
A practical guide to incident response and digital forensics
Where SOAR playbooks fit and where AI agents take over
Free ebook: modern security operations with Microsoft Sentinel
How the framework is structured and what changes from the Essential Eight
How an attack unfolds stage by stage and how to detect each one
November 16, 2026
November 13, 2026
November 10, 2026
We provide end-to-end SecOps solutions for Managed Security Service Providers (MSSPs), Managed Service Providers (MSPs), and Value-Added Resellers (VARs).
Find out how you can become a ThreatDefence partner.
Protect your entire technology stack and build your SecOps on your own pace
White-labeled AI SOC for every client tenant, co-managed with our SOC team
Launch your SOC business without any upfront investment
Embed security operations, threat intelligence and exposure data in your product
Sovereign SOC for Australian MSPs and MSSPs, with onshore analysts and client data
Set the rules for closing alerts automatically, with evidence checks and review
How MSPs deliver branded security monitoring with a SOC partner
ThreatDefence was an official partner of the Cyber Resilience Summit NSW, 29 October 2026 at Hilton Sydney.
Our team was back at BSides Canberra, and our cyber quiz was a hit again.
Two days in Perth with our partner Ever Nimble, on security operations and cyber resilience.
Sponsors of the AUSCERT Cyber Security Conference on the Gold Coast.
Sponsors of the Gartner Security & Risk Management Summit in March.
Sponsors of the Cyber Security for Critical Assets Summit in Perth, February 2026.
Expert conversations with speakers and attendees at CyberCon Melbourne 2025.
Credential stuffing and password spraying are attacks that guess their way into accounts at scale. Credential stuffing tries username and password pairs stolen from other services. Password spraying tries a few common passwords against many accounts.Both leave a recognizable shape in sign-in data. This guide puts the two side by side, shows what each looks like in the logs, and explains how to read the result of every attempt.
Brute force, by contrast, tries many passwords against one account. It triggers lockouts quickly and is easy to see. Spraying and stuffing exist because they avoid that.
Both attacks run on lists, and the lists are cheap to build, which is why almost every organization is a target, whatever its size.
An attacker needs valid usernames before any guessing starts. For most organizations these are easy to produce. Staff names are public on professional networking sites and company pages. The email format, such as first initial and surname, can be read from a single published address. Combining the two gives a list that is mostly right.
Some sign-in pages confirm whether a username exists, through a different error message or response time. Attackers use this to clean their lists before trying any passwords, which is why accounts that do not exist appear early in an attack and disappear later.
Spraying needs a short list of passwords that someone in the organization is likely to have chosen. The candidates are well known: the season or month with the year, the company name with a number, the name of the local sports team, the word welcome with a digit, and the default the help desk hands out to new hires.
Password rules that force a change every 90 days make this easier. People who must invent a new password four times a year fall back on patterns, and the patterns follow the calendar. Attackers time their rounds to follow expiry dates, when many people have just chosen something easy to remember.
Stuffing uses real credentials. Some come from breaches of other services, where email addresses and passwords were stolen and later shared in large compilations. Others come from infostealer malware, which collects saved passwords directly from infected computers, as the infostealer guide describes.
The lists are enormous, traded freely, and updated constantly. An attacker filters one for addresses at the target’s domain and has a ready-made set of username and password pairs. Any employee who used their work email and a favorite password to register for an outside service is on it.
This is why password strength alone is no defense against stuffing. A long, complex password that was used on a breached site is exactly as exposed as a short one.
Neither attack is typed by hand. Both are run with automation tools that take a list, a target, and a set of proxy addresses.
Attackers rent access to pools of addresses, including residential connections that belong to ordinary households. Each attempt can come from a different address in the right country, which defeats blocking by address and makes the traffic look like staff signing in from home.
The tools throttle attempts to stay under lockout thresholds and rate limits, and spread a spray across days.
Attackers prefer endpoints with the weakest checks: password-only mail protocols where they are still enabled, programmatic interfaces, and sign-in pages for secondary applications that lack the protections of the main portal.
Successful pairs are logged, checked for a second factor, and sorted by the access they give. Accounts without a second factor are used at once, and the rest are sold or held for a follow-up attack.
A campaign rarely targets one organization. The same tool runs the same list against thousands of tenants. Being hit says nothing about whether an organization was singled out.
A spray is slow and even. The same source works through the directory, one attempt per account, then waits and starts again with the next password.
Several features identify it. Accounts are tried in alphabetical order. Each account fails once per round. Accounts that no longer exist are tried, because the list was built from old public data. The gap between rounds, 45 minutes here, keeps each account under the lockout threshold.
The fifth row matters most. The password for that account was correct, and only the second factor stopped the sign-in.
Stuffing is fast and scattered. Each attempt comes from a different address, and each account is tried with a password its owner really used somewhere.
The sources are spread across many networks, often home broadband addresses rented through proxy services. Addresses in a format the organization does not use for sign-in appear, such as old aliases and departed staff, because the list came from other services’ breaches and not from the directory. Among the failures there are a few successes.
A tenant’s sign-in log holds only attempts against its own domains. The personal addresses in the same breach list are tried against consumer services and never appear in it.
Identity platforms record why each sign-in failed. The reasons mean different things, and the correct-password failures are routinely misread as harmless.
In Microsoft Entra ID sign-in logs these results carry error codes. A wrong password is 50126, a username that does not exist is 50034, and a smart lockout is 50053. A correct password that stopped at the second factor is 50074 or 50076, and one blocked by an access policy is 53003.
A sign-in that failed at the second factor still proves the password is known. Left alone, that account is the next target for repeated prompts, a help desk call, or a phishing page, and the MFA fatigue guide follows that next step.
Attackers design around lockout rules, so it helps to know what those rules do.
Microsoft Entra ID uses smart lockout. By default it locks an account after 10 failed attempts, for one minute at first and longer after further failures. It keeps separate counts for familiar and unfamiliar locations, so an attacker’s failures from a new location lock the attacker out while the real user continues to sign in. It also ignores repeated attempts with the same wrong password.
A spray that tries one password per account every 45 minutes never approaches that threshold, which is why lockout alerts rarely reveal a spray.
On-premises directories use fixed lockout policies. A low threshold stops guessing and also lets an attacker lock out every employee on purpose. Remote access devices that authenticate against the directory inherit the same weakness.
One address fails against many different accounts within an hour. This is the core spray detection, because it counts accounts per source where lockout counts failures per account.
Each account shows roughly the same number of failures, which human error never produces.
A burst of sign-ins arrives for usernames that are not in the directory.
The same user agent or client application appears across attempts for many users.
Failed sign-ins rise suddenly across the tenant from a large number of addresses, which is the shape of stuffing.
Any successful sign-in from a source, network, or client seen in the attack needs review.
Sign-ins stopped by the second factor or an access policy are listed and reviewed daily.
Attempts arrive against SMTP AUTH, on-premises Exchange, or another mail system that accepts a password alone.
Microsoft Entra ID Protection raises a password spray detection, and flags leaked credentials when it finds a user’s password in breach data.
Attention goes to the main cloud portal, which is usually the best protected entry point. The same lists are tried everywhere a directory password is accepted.
These often authenticate against the internal directory and may not enforce a second factor for every account. Their logs record failed logons by username and source, and are rarely reviewed.
These carry the same exposure, with direct access to internal systems on success.
Customer portals, supplier portals, and older internal applications published to the internet keep their own password checks.
Basic authentication is switched off in Exchange Online for every protocol except SMTP AUTH. The exposure that remains is SMTP AUTH, on-premises Exchange, and non-Microsoft mail systems.
An attacker who already has a foothold can spray the directory from inside, where there is no second factor at all.
Internal spraying looks different in the logs. Domain controllers record failed Kerberos pre-authentication (event 4771) and failed logons (event 4625). A spray appears as one source machine generating failures for many accounts in a short time, each account failing once.
Attackers on the inside also spray local administrator accounts across workstations, looking for a shared password. One machine attempting to log on to many others with the same account name is the signature.
Organizations that run their own customer portal, online store, or member site face credential stuffing from the other side. The accounts at risk belong to customers, and the organization is responsible for protecting them.
The attack is the same. Lists of email addresses and passwords from other breaches are tried against the login page, and the pairs that work are used to take over customer accounts, steal stored payment details or loyalty points, or commit fraud.
Controls for a customer login differ from those for staff, because customers cannot be forced onto managed devices.
Attempts are restricted per account and per source, and responses slow down after failures.
Traffic that does not behave like a browser operated by a person is identified and challenged.
At registration, login, and password change, the password is compared against known breached passwords, and a match requires a change.
It is offered to all customers and required again before sensitive actions such as changing payout details or delivery addresses, even within a signed-in session.
Customers are told about sign-ins from new devices and about changes to their account.
The ratio of failed to successful logins is watched, since a sudden rise in failures across many accounts is the same signal as in a staff tenant.
Where the attacker reaches personal information in customer accounts, the organization has to assess the incident as a data breach, even though the passwords were leaked elsewhere. Regulators have taken the view that an organization should anticipate credential stuffing and defend against it.
Much traditional password policy makes these attacks easier. Current guidance from the US National Institute of Standards and Technology, in its digital identity guidelines, points the other way.
Encourage long passwords and passphrases. Rules that demand symbols and mixed case produce predictable substitutions.
Change passwords when there is evidence of compromise, and drop the calendar.
Check new passwords against lists of commonly used and previously breached passwords, and reject matches.
Add the organization’s name, products, locations, and help desk defaults to the blocked list. In Microsoft Entra ID the custom banned password list holds up to 1,000 terms and also blocks common variations of each.
New-hire and reset passwords should be random, single-use, and expire on first sign-in.
It is the practical way to make unique passwords for outside services achievable.
These measures reduce the hit rate. They do not remove the need for a second factor.
These checks show how an attack would go before one happens.
Multi-factor authentication on every account is the control that turns both attacks from a breach into an alert. Phishing-resistant methods go further where password sign-in is turned off for the account, because there is then no password to guess.
ThreatDefence watches Microsoft 365 and Entra ID sign-in failures for the patterns in this guide, and for the successful sign-in that follows them. AI agents review each alert against the account’s sign-in history, and our SOC locks a compromised account within minutes of a confirmed sign-in.
The platform also monitors leaked credentials for customer domains, so a password that appears in a breach can be reset early. A free Microsoft 365 forensic review reports password spray activity, MFA coverage, and risky sign-ins over the past six months.
Experiencing a breach?
Our incident response team answers 24x7, with or without a retainer.
Explore Identity Threat Detection and Response
Ask about a free Microsoft 365 forensic review, covering six months of activity in your tenant.
CountryAustraliaUnited StatesNew ZealandUnited KingdomCanadaSingaporeOtherI'm an MSP, MSSP, or reseller
Δ
Credential stuffing tests username and password pairs already exposed at other services, so it succeeds where someone reused a password. Password spraying tries one or two common passwords against every account, so it succeeds where someone chose a predictable one. In the logs, stuffing is fast and comes from thousands of addresses, while a spray is slow, even, and often from a single source.
A brute force attack tries many passwords against one account. It is the loudest of the password attacks: failures pile up on a single account, lockout triggers within minutes, and the pattern is plain in the log. Spraying and stuffing are the quiet alternatives, built to spread attempts across accounts or sources so that no single counter is tripped.
Lockout counts failures per account, and a spray produces one failure per account per round. Smart lockout in Microsoft Entra ID triggers after 10 failed attempts by default, so a spray that tries one password every 45 minutes never reaches it. Detection has to count the other way: one source failing once against many different accounts within an hour.
It stops the sign-in, and the password is still compromised. The log shows the attempt as a correct password that failed at the second factor, and that account becomes a target for repeated prompts, a help desk call, or a relay phishing page. Reset the password. Check too for routes that skip the second factor, such as SMTP AUTH, a VPN, or accounts excluded from policy.
List every sign-in from the attacking sources where the password was accepted. In Microsoft Entra ID that means successes, plus failures with error 50074 or 50076, where the password was right and the second factor stopped it, and 53003, where an access policy did. Then search for later successful sign-ins to those accounts from any new address, since attackers return through a different proxy.
No. Disabling every account that received a failed attempt does the attacker’s work, since it stops the business over passwords that were wrong anyway. Act only on accounts where the password was correct: reset them, and block sign-in for any that show a success. Blocking the source addresses helps against a single-source spray and does little against thousands of rotating proxies.
The attempts alone are not. It becomes a breach question when a sign-in succeeded and the account held or could reach personal information. For staff accounts that means the mailbox and files. For customer portals it means each customer account the attacker opened, and regulators have taken the view that organizations are responsible for those even though the passwords were leaked by another service.
They come from breach compilations and from infostealer logs. Breach compilations hold email addresses and passwords taken from compromised websites, often years old. Infostealer logs are fresher and more dangerous, because the malware copies every password saved in a browser along with the address of the site it belongs to. An attacker filters either source for the target’s email domain and has a list in minutes.
Account Takeover, Followed Through the Logs
MFA Fatigue Attacks
Infostealer Malware