Account Takeover Prevention: How to Detect and Stop ATO Fraud
Introduction
Account takeover rarely starts with a dramatic hack. It usually starts with a valid username and password that someone else bought, phished, or guessed. The login looks normal. The session looks normal. Then the shipping address changes, the gift card balance disappears, or a payout goes to a new bank account.
That is why account takeover fraud is so hard to stop with passwords alone. The attacker is not breaking in. They are logging in.
The trend is moving in the wrong direction for businesses that rely on customer accounts.
LexisNexis Risk Solutions reported that account takeover attempts soared in 2025 after a brief decline in 2024, with the overall account takeover attack rate reaching 1.2%, up 89% year over year. In ecommerce, the login attack rate jumped to 10.8% from 3.4% the year before. Source: LexisNexis Risk Solutions, Cybercrime Report 2026
This guide explains what account takeover is, how ATO attacks work, and which signals help you detect them early. It also shows how to add account takeover prevention to login, account changes, and checkout. Greip supplies IP, network, and identity risk signals. Your own rules decide when to allow, challenge, or block.
What Is Account Takeover (ATO) Fraud?
Account takeover (ATO) fraud happens when a criminal gains control of a real customer's account and uses it for their own benefit. The account belongs to a genuine person, which is what makes it valuable.
Once inside, attackers typically:
- Spend stored balances, loyalty points, or gift cards
- Place orders with saved payment cards
- Change the email, phone number, or password to lock the owner out
- Add a new payout or bank account and move money out
- Harvest personal data to sell or reuse in other fraud
ATO is different from new account fraud. With new account fraud, the identity is fake or stolen from the start. With account takeover, the account has history and trust, so many checks wave it through.
Why Account Takeover Hurts More Than a Single Bad Transaction
A takeover is not one loss. It is a chain of costs.
There is the direct loss from purchases or withdrawals. There are chargebacks and refunds when the real customer disputes them. There is support time spent restoring access and checking what changed.
Then comes the trust damage. A customer who loses money or data through your platform often leaves, even if you refund them. Sift's Q3 2025 account takeover index notes that losses from ATO fraud are projected to climb to $17 billion in 2025, up from $13 billion the year before.
The industries hit hardest tend to hold stored value or reusable data:
- Online retail and marketplaces with saved cards and loyalty points
- Fintech, banking, and wallets with direct access to funds
- Travel and ticketing with points and stored payment methods
- Gaming platforms with in-game currency and tradable items
- Streaming and SaaS with subscriptions that can be resold
How Account Takeover Attacks Happen
Most ATO attacks follow the same basic path: get credentials, test them, take control, cash out.
Getting Credentials
Attackers collect logins in a few common ways:
- Data breaches: leaked username and password lists from other sites
- Phishing: fake login pages sent by email, SMS, or chat
- Infostealer malware: software on the victim's device that captures saved passwords and session cookies
- Social engineering: tricking support teams or users into resetting access
Testing Credentials at Scale
Leaked credentials are tested with bots. This is credential stuffing. Because people reuse passwords, even a small match rate produces many working accounts. Our guide to credential stuffing attacks covers this stage in depth.
Taking Control and Cashing Out
After a successful login, attackers move fast. They change contact details, add new devices, or update payout information. Then they spend, withdraw, or resell the account before the owner notices.
The Warning Signs of an Account Takeover Attempt
ATO detection works best when you watch for change. A real customer tends to log in from similar places, networks, and devices. An attacker breaks that pattern.
Common warning signs include:
- A login from a new country far from the user's usual location
- Two logins from distant places within a short time (often called impossible travel)
- Logins through a VPN, proxy, Tor, or hosting network when the user normally uses a home or mobile connection
- Many failed logins across different accounts from the same IP or network
- A password reset followed quickly by an email or phone change
- A new payout account or shipping address right after a risky login
- A high-value order shortly after a contact detail change
No single signal proves fraud. Plenty of real customers travel or use a VPN. The goal is to score each event and respond in proportion to the risk.
Why Passwords and MFA Alone Don't Stop ATO
Strong passwords help, but they do not fix reused or stolen credentials. Multi-factor authentication helps a lot, yet it has gaps too.
One-time codes can be phished in real time. SIM swaps can redirect SMS codes. Stolen session cookies can skip the login step entirely. And if you prompt for MFA on every login, users get frustrated and some turn it off.
A better approach is risk-based authentication. You let low-risk logins through smoothly and step up only when the signals look wrong. Our article on risk-based MFA explains how to decide when to challenge.
Using IP and Network Intelligence to Detect Account Takeover
The IP address behind a login is one of the fastest signals you can check. It tells you where the request comes from and what kind of network sent it.
Checking Location and Network on Each Login
Greip's IP Lookup endpoint returns location fields such as country_code, city, latitude, and longitude in the geo object. When you pass params=security, it also returns a security object. You call it from your backend with your API key in the Authorization header.
curl --request GET \
--url 'https://v2.greipapi.com/lookup/ip/165.227.149.217?params=security' \
--header 'Authorization: Bearer <token>'The security object in the documented sample response looks like this:
"security": {
"vpn_or_proxy": true,
"vpn_or_proxy_type": "VPN",
"tor": false,
"bot": false,
"relay": false,
"relay_country": null,
"relay_city": null,
"hosting": true,
"is_blacklisted": false
}For ATO, a few fields matter most. vpn_or_proxy and tor show that the real location is hidden. hosting shows the request comes from a cloud or data center network, which is unusual for a normal customer login. is_blacklisted shows the IP appears on a threat list. The asn object, including its type, tells you which network operator owns the IP.
You store the country and coordinates from each successful login. On the next login, you compare them. A sudden jump to a new country, or a distance that no one could travel in the time between logins, is a strong reason to step up verification. Greip provides the location data. Your system keeps the history and does the comparison.
Pass the user_id query parameter (for example, your internal user id) on lookups. Greip's user identifier option lets you trace risky events back to a specific account on the Events page in the Greip dashboard.
Getting a Single IP Risk Score
If you prefer one number, Greip's IP Scoring endpoint at GET /score/ip/{ip} returns a score from 0 to 100, a risk_level of low, medium, or high, and a recommendation of accept, review, or reject. Here is the documented sample:
{
"success": true,
"data": {
"ip": "12.12.12.12",
"score": 35,
"risk_level": "medium",
"recommendation": "review",
"rules": [
{
"id": "IP_VPN_OR_PROXY",
"description": "IP address is associated with VPN, proxy, or anonymization service."
},
{
"id": "IP_HOSTING",
"description": "IP address belongs to hosting or cloud provider ASN."
}
]
}
}The rules array lists only the signals that fired, so your logs show why a login was flagged. That makes reviews and customer support conversations much easier.
Protecting Account Changes, Not Just Logins
Many teams protect the login page and forget what happens next. Attackers know this. The most damaging actions come after login.
Treat these events as high risk and check them again:
- Password resets and account recovery
- Email or phone number changes
- New payout, bank, or withdrawal details
- New shipping addresses on high-value orders
- Redeeming points or gift card balances
When a user changes their email, you can score the new address with Greip's email scoring. The response includes fields such as is_disposable, risk_level, and recommendation, which help you spot a switch to a throwaway inbox. When a user adds a phone number, phone scoring returns is_valid, is_disposable, and a risk_level you can use the same way. Both are part of Greip's data scoring and validation service.
Do not rely on the new email or phone to confirm a contact change. Send the confirmation to the old, trusted contact as well, so the real owner can stop the change.
How to Build an Account Takeover Prevention Flow
You do not need a huge project to start. A simple, layered flow covers most risk.
Step 1: Collect Signals at Login
On every login attempt, look up the IP. Record the country, city, coordinates, ASN, and security flags. Count failed attempts per account and per IP over short windows.
Step 2: Compare Against the User's History
Check whether the country, network type, or device is new for this user. Calculate travel distance and time since the last trusted login. Flag VPN, proxy, Tor, and hosting traffic when it is unusual for that account.
Step 3: Map Risk to Clear Actions
Keep the responses simple and predictable:
- Low risk: allow the login with no friction
- Medium risk: step up with MFA or an email confirmation
- High risk: block the attempt, lock sensitive actions, and alert the user
Step 4: Re-check Sensitive Actions
Run the same checks before contact changes, payout changes, and large orders. A session that started with a risky login should face extra checks for the rest of its life.
Greip's account takeover use case page and the account takeover integration guide show where these API calls fit.
Real-World ATO Scenarios
A Retail Account Drained of Loyalty Points
A customer usually logs in from a home connection in Texas. One night, a login succeeds from a hosting IP in another country. Minutes later, someone tries to redeem all their points. A rule that requires step-up verification for logins from hosting networks, plus a new-country check, stops the redemption.
A Wallet With a New Payout Account
A fintech user's account resets its password, changes its email, and adds a new bank account within ten minutes. The reset came from an IP flagged as vpn_or_proxy. Holding the payout change for review and notifying the old email address gives the real owner time to respond.
A Gaming Platform Under Credential Stuffing
Thousands of failed logins hit different usernames from a small set of IPs in the same ASN. Rate limits by IP and ASN, combined with a high IP risk score, slow the attack down while real players log in as usual.
Common Challenges and How to Handle Them
False Positives on Travelers and VPN Users
Real people travel and use privacy tools. Do not block on one signal. Use step-up verification instead, and let users confirm new locations easily.
Residential Proxies
Some attackers route traffic through residential IPs to look like home users. Combine IP signals with velocity, account history, and contact change patterns, so you are not relying on network type alone.
Shared and Mobile IPs
Mobile carriers and offices share IPs across many people. Rate limit carefully and weigh account-level signals more than IP-level counts on these networks.
Session Hijacking
If an attacker steals a session cookie, they may never hit your login page. Re-checking the IP and location on sensitive actions catches sessions that suddenly move to a new network.
Account Takeover Prevention Best Practices
- Score every login, not only failed ones
- Use risk-based MFA instead of prompting everyone every time
- Re-verify before contact, password, and payout changes
- Notify users on the old contact details when something changes
- Rate limit by IP, ASN, and account
- Watch for anonymized traffic with VPN and proxy detection
- Log the reasons behind every decision so support can explain them
- Review false positives monthly and tune your thresholds
ATO Trends to Watch
Attackers are automating more of the work. Bots test credentials faster, and infostealer malware delivers fresh logins and session cookies at scale. Fraud teams are also seeing more attacks that mix social engineering with technical access.
Passkeys and phishing-resistant authentication are spreading, which raises the bar for simple credential attacks. Even so, account recovery flows and sessions remain targets. Expect continuous, risk-based checks across the whole session to become the norm, rather than one check at the front door.
Frequently Asked Questions
What is account takeover fraud?
Account takeover fraud is when a criminal gains control of a real customer's account, usually with stolen or phished credentials, and uses it to steal money, make purchases, or harvest personal data.
How do account takeover attacks happen?
Attackers collect credentials through data breaches, phishing, or malware, test them at scale with bots (credential stuffing), then log in, change contact details, and cash out before the owner notices.
How can you detect an account takeover attempt?
Watch for changes from the user's normal pattern: new countries, impossible travel, VPN, proxy, Tor, or hosting networks, bursts of failed logins, and quick changes to email, phone, password, or payout details.
Is MFA enough to prevent account takeover?
MFA helps a lot, but it is not enough on its own. One-time codes can be phished, SMS can be redirected by SIM swaps, and stolen session cookies can skip login. Risk-based checks on logins and sensitive actions close those gaps.
What is the difference between account takeover and credential stuffing?
Credential stuffing is a method: bots test leaked username and password pairs across many accounts. Account takeover is the outcome: an attacker controls a real account, whether through credential stuffing, phishing, or other means.
How does Greip help with account takeover prevention?
Greip provides risk signals through its API, including IP location, VPN, proxy, Tor, and hosting detection, IP risk scoring, and email and phone scoring. You use these signals in your own rules to allow, challenge, or block logins and account changes.
Conclusion
Account takeover is a login problem and a trust problem at the same time. Attackers use real credentials, so the only reliable way to catch them is to notice when something changes.
Start with the signals you can check on every request: where the login comes from, what network sent it, and whether contact or payout details are changing. Map that risk to clear actions, from no friction to step-up verification to a block. Greip gives you the IP, network, and identity data. Your rules keep real customers moving and keep attackers out.
Get started
Start protecting your business today
Our service is trusted by thousands of businesses worldwide.
- 1,000 requests during trial
- Cancel anytime