Hooded figure sitting at a laptop in a dark room lit in teal light
Cybersecurity

Credential Stuffing: Why Online Stores Get Hit and How to Stop It

A credential stuffing attack uses passwords leaked elsewhere to break into your customers' accounts. How it works, why stores are targets, and the defences that work.

F

Force Media Team

7 min read

Photo: Clint Patterson on Unsplash

Nobody broke into your website. Nobody found a bug in your code. And yet, on Monday morning, three customers email to say someone ordered from their accounts using their saved cards, and a fourth says her loyalty points have vanished.

That is what a credential stuffing attack usually looks like from the inside. The attacker didn't hack you. They hacked someone else years ago, took the list of emails and passwords, and tried every one of them on your login page, betting that some of your customers use the same password everywhere. Some always do.

How credential stuffing works

The mechanics are almost boringly simple, which is exactly why it's so common.

Anatomy of a credential stuffing attack

  1. 01

    Leak

    Another site is breached; emails and passwords spill out

  2. 02

    Collect

    Lists are combined and traded or sold

  3. 03

    Automate

    Bots replay every pair against your login form

  4. 04

    Hit

    A small share works because people reuse passwords

  5. 05

    Cash out

    Orders, gift cards, points, stored data, resale of accounts

OWASP draws a useful line between this and its cousin, password spraying. Credential stuffing replays specific username-and-password pairs leaked from another site. Password spraying tries one common password (think Summer2026!) against many accounts (OWASP). The defences overlap, but stuffing is the one that punishes password reuse.

How common is it, really?

More common than most business owners assume. For its 2025 Data Breach Investigations Report, Verizon looked at two years of single sign-on logs and found that credential stuffing was a steady background hum on almost every login system.

Share of daily login attempts that were credential stuffing

Median daily share of authentication attempts identified as credential stuffing, over two years of SSO logs. Source: Verizon, 2025 DBIR research on credential stuffing

The same research found compromised credentials were the way in for 22% of breaches that year, and that people's passwords are depressingly similar across services: on infected devices, only about half (49%) of a user's saved passwords were different from each other (Verizon).

A real example: 14,000 logins, 6.9 million people

The case that made credential stuffing mainstream news was 23andMe in 2023. According to the company's amended SEC filing, attackers used passwords leaked elsewhere to get into about 0.1% of accounts, roughly 14,000. Because those accounts could see relatives through the DNA Relatives feature, the data exposed reached about 6.9 million people. 23andMe said it found no breach of its own systems (HealthITSecurity).

"Nobody hacked our servers" is true and beside the point. If attackers can log in as your customers, the damage is yours to deal with.

The lesson from 23andMe

That's the uncomfortable part for any online store: the weakness lives in your customers' habits, but the fallout, from refunds and chargebacks to support time and lost trust, lands on you.

Why online stores are attractive targets

Online stores combine everything an attacker wants behind one login:

  • Saved payment methods that allow a quick purchase without re-entering a card.
  • Stored value: gift card balances, loyalty points, store credit.
  • Personal data: addresses and phone numbers that feed further scams.
  • Accounts as goods: working logins for popular shops are themselves resold.

And unlike a bank, a small shop often has a plain email-and-password form with no rate limiting at all. To a bot, that's an open door. Forget the cliché of a lone hacker in a dark room, too: the reality is usually a rented bot running a purchased list, thousands of attempts an hour.

Padlock resting on a backlit laptop keyboard glowing in red and green light
A lock only helps if the key isn't already in circulation. Stolen passwords are keys that work on many doors. Photo: FlyD on Unsplash

The defences that actually work

No single control stops credential stuffing. Layers do. Here's what we put in place, roughly in order of impact for a small or mid-sized store.

  1. 1

    Offer multi-factor authentication, and nudge people into it

    Microsoft's widely quoted 2019 finding was that MFA 'can block over 99.9 percent of account compromise attacks'. Verizon's 2025 research is more cautious, noting MFA is 'not the panacea it used to be' as attackers find ways around it, but its absence is still one of the two main weaknesses behind stuffing. Passkeys are even better where your platform supports them.

  2. 2

    Block passwords that are already leaked

    NIST's digital identity guidelines (SP 800-63B) say services should 'compare the prospective secret against a blocklist that contains known commonly used, expected, or compromised passwords'. Do it at sign-up and at password change.

  3. 3

    Rate-limit and slow down failed logins

    Limit attempts per account and per IP, add progressive delays, and show a challenge after repeated failures. NIST caps consecutive failures at 'no more than 100' per account; most stores should be far stricter.

  4. 4

    Look at signals, not just passwords

    New device, new country, a login followed instantly by a change of email or address: these patterns deserve a step-up check or a hold.

  5. 5

    Tell customers when something unusual happens

    A 'new sign-in from Chrome on Windows' email gives customers a chance to stop an attacker before an order ships.

  6. 6

    Protect the actions that matter

    Ask for the password again (or a one-time code) before changing the email, adding a delivery address or spending stored credit.

Checking for leaked passwords, without seeing them

Blocking breached passwords sounds hard, but it isn't. Have I Been Pwned's free Pwned Passwords service uses a "k-anonymity" range lookup: you send only the first five characters of the password's SHA-1 hash, get back every matching suffix, and check locally. The password itself never leaves your server.

import crypto from 'node:crypto'

// Returns how many times a password appears in known breaches (0 = not found).
export async function timesPwned(password) {
  const hash = crypto.createHash('sha1').update(password).digest('hex').toUpperCase()
  const prefix = hash.slice(0, 5)
  const suffix = hash.slice(5)

  const res = await fetch(`https://api.pwnedpasswords.com/range/${prefix}`)
  const lines = (await res.text()).split('\n')

  const match = lines.find((line) => line.startsWith(suffix))
  return match ? parseInt(match.split(':')[1], 10) : 0
}

// At sign-up or password change:
// if (await timesPwned(newPassword) > 0) reject with a friendly message.

What to do if you think it's happening now

First hour response

  • Check your login logs for a spike in failures, especially from many IPs against many accounts
  • Turn on (or tighten) rate limiting and bot challenges on the login and password-reset pages
  • Force a password reset for accounts that logged in successfully from suspicious sources
  • Pause stored-value spending and address changes on affected accounts
  • Tell affected customers plainly what happened and what you've done
  • Write down the timeline; you may need it for your payment provider or regulator

If your store runs on WordPress, a lot of this is plugin territory, and plugins bring their own risks; our guide to WordPress plugin security covers how to choose them. For the broader picture, the website security essentials checklist is a good companion to this post.

Not sure how your login page would hold up? Ask us for a security review. We'll test it the way an attacker would and tell you plainly what to fix first.

Speak to our experts

From AI agents to secure, fast websites, we build technology that does real work for your business.

Get in touch
FAQ

Common Questions

Everything you need to know about working with Force Media.

Ask Us Anything
We primarily serve businesses in Sri Lanka and Australia, but our remote-first approach means we can work with clients globally.
Project timelines vary by scope. A standard website takes 2–4 weeks. Custom software projects range from 6–16 weeks depending on complexity.
Yes. All our projects include a support period, and we offer ongoing retainer packages for maintenance, SEO, and feature development.
Absolutely. We specialize in integrating with existing platforms, APIs, and legacy systems to extend their capabilities.