# Fake account detection that stops bot signups at the front door

Source: https://kavralab.com/solutions/fake-accounts/

**Fake account creation** is the bulk registration of accounts that do not belong to real, distinct customers, made by bots, device farms or synthetic identities and later used for spam, promo abuse, fake reviews, mule activity or resale. Kavra assesses each signup for automation, spoofed devices and shared infrastructure before the account exists.

- **Who it hits:** Marketplaces, social, fintech, SaaS, retail
- **What it costs:** Spam, fraud, SMS fees, polluted metrics
- **Tools used:** Signup bots, emulators, device farms, proxies
- **Where to stop it:** At signup, before the account is live

## What is fake account creation?

A fake account is any account registered without a real, distinct person or business behind it who intends to use the service as designed. Some are created by scripts in the thousands. Some are created by hand in phone farms. Some carry a convincing but invented identity. What they share is that the account is a tool for something else.

Fake accounts are the raw material of most online abuse. They are cheap to make in bulk, stored until needed, and often sold as aged accounts that look more trustworthy than fresh ones. Stopping them at signup is far cheaper than finding them later by what they do.

## What fake accounts are used for

The signup is rarely the goal. Knowing the end use tells you where the damage will land.

| Use | What happens next | Who feels it |
|---|---|---|
| Promo and referral farming | Welcome offers and referral rewards claimed once per fake account | Growth and finance teams |
| Spam and scams | Messages, comments and fake listings pushed to real users | Trust and safety, users |
| Fake reviews and ratings | Products, sellers or apps boosted or attacked | [Marketplaces](https://kavralab.com/industries/marketplaces/) and their buyers |
| Mule and payment fraud | Accounts receive and move stolen funds or test stolen cards | Fintech compliance and payments |
| SMS and OTP abuse | Signups trigger verification texts to premium numbers | Telecom budget, see [SMS pumping](https://kavralab.com/solutions/sms-pumping/) |
| Free tier and API farming | Trials, credits and API keys multiplied | SaaS and AI product costs |
| Resale | Aged or verified accounts sold to other fraudsters | Every team, later |

## How bulk fake account creation works

Fake account operations run like small factories, with each stage bought or automated.

1. **Source identities**: Emails come from disposable or bulk-registered domains, phone numbers from virtual number services, and names from generators or leaked data. For regulated products, [synthetic identities](https://kavralab.com/glossary/synthetic-identity/) mix real and invented details to pass onboarding checks.
2. **Provision devices**: Scripts run in headless browsers on servers, Android [emulators](https://kavralab.com/detect/emulators/) imitate phones, and [device farms](https://kavralab.com/detect/device-farms/) run racks of real handsets for apps that check hardware.
3. **Rotate networks**: Each signup exits through a different residential or mobile proxy so that no single IP address stands out.
4. **Automate the form**: Bots fill the signup form, click email links, read codes from virtual numbers and pass CAPTCHAs through solving services. Human workers handle the steps that bots fail.
5. **Age and deploy**: Accounts are left to sit, warmed up with light activity, then used in a campaign or sold. Aged accounts slip past rules that only distrust brand-new users.

## Who it hits and what it costs

Every product with open registration gets fake signups. The cost depends on what a new account can do. On marketplaces and social products, fake accounts erode trust in listings, reviews and messages. In fintech, they open the door to mule networks and regulatory scrutiny. In SaaS and AI, each fake account consumes compute, credits and support time. In retail and iGaming, they turn marketing budgets into payouts for fraud rings.

- Direct fraud: rewards, credits and payments captured through accounts that should not exist.
- Variable costs: SMS verification, email sending, compute and storage for accounts with no customer behind them.
- Bad data: inflated signup, activation and user counts that mislead planning and investors.
- Cleanup: moderation, bans and appeals for accounts that could have been refused in a single step.

## Warning signs of fake signups

Fake accounts give themselves away by what they share and how they are made.

- **Signup surges**: Sudden spikes of new accounts with no matching campaign, often at night in your main market.
- **Emulated or virtual devices**: Phones that run on desktop hardware and browsers inside virtual machines.
- **Automation traces**: Headless browsers, automation frameworks and forms completed with no real typing.
- **Patterned emails**: Disposable domains, sequential usernames or random strings on free providers.
- **Shared infrastructure**: Many signups sharing a device, proxy provider or network range.
- **Dormant after signup**: Accounts that verify, then go quiet until a campaign or a sale.

## Why common signup defenses fall short

Signup checks often validate the details, not the actor typing them.

| Defense | What it proves | What it misses |
|---|---|---|
| Email verification | The inbox exists | Disposable and bulk-registered inboxes verify fine |
| SMS verification | The number receives texts | Virtual numbers, and each text costs you money |
| CAPTCHA | Someone solved a puzzle | Solving services and farm workers solve them at scale |
| IP rate limits | Few signups per address | Proxy pools give every signup its own address |
| Document checks | An ID was presented | Synthetic and stolen identities, plus friction for real customers |

## How to prevent fake account creation

The best time to stop a fake account is before it exists. That needs a decision at the signup request, based on who is really submitting it, with extra checks saved for the uncertain middle.

- Assess every signup on web, mobile app and API, not only the website form.
- Detect automation, emulators, virtual machines and device farms directly.
- Link signups by device and network so a surge from one operator shows up as one campaign.
- Recognize residential and mobile proxy exits even when the IP looks like a home connection.
- Send SMS codes only after the signup passes a risk check, to avoid paying for fake verifications.
- Verify or limit uncertain new accounts, for example by delaying rewards or messaging, instead of rejecting them outright.
- Recheck accounts when they first do something valuable, so aged fakes are caught too.

Measure the result by what fake accounts stop doing, not only by how many signups you reject. Fewer spam reports, lower SMS spend per real customer, fewer referral payouts to linked accounts and cleaner activation numbers are the signs the filter is working. If rejections rise but downstream abuse does not fall, the fakes are getting through a door you are not watching, such as the mobile app or a partner API.

## A real signup vs a fake account

**Real signup**

- A real phone or computer that behaves like one
- Home or carrier network that fits the device
- Typing, pauses and the odd mistake
- Uses the product after signing up

**Fake account**

- Emulator, virtual machine or farm handset
- Proxy exit, new IP for every signup
- Form filled by script or at machine speed
- Idle, then used for one job

> **Key takeaway:** Fake accounts are cheap to make and expensive to clean up. Judge the actor behind each signup, not just the details it types, and stop the factory at the front door. The same signals help against [multi-accounting](https://kavralab.com/solutions/multi-accounting/) and [bonus abuse](https://kavralab.com/solutions/bonus-abuse/).

## How Kavra stops fake account creation

Kavra assesses each signup with 3,000+ data points and returns a clear recommendation before your backend creates the account.

- **Signup bots exposed**: Headless browsers, automation frameworks and HTTP clients imitating browsers are caught by contradictions between layers.
- **Emulators and farms**: Emulators, virtual machines, device farms and spoofed devices are flagged on web and through native iOS and Android SDKs.
- **Coordinated campaigns**: Surges of new devices, shared infrastructure and velocity anomalies are grouped, so one operator's burst shows up as one campaign.
- **Proxy intelligence**: Kavra measures real exit IPs of commercial residential and mobile proxy networks, on top of 30+ public reputation feeds.
- **Invisible to real users**: No CAPTCHA, puzzle or slider, ever. Extra invisible checks run when evidence is unclear, with nothing for the visitor to solve.
- **Your policy, your call**: Choose cautious, balanced or strict presets, start in observe-only mode, and let your backend allow, verify or reject.

## FAQ

### How do you detect fake accounts?

Look at how the account was made, not only at the details entered. Automation, emulated devices, proxy networks and many signups sharing one device or network are strong signs. Kavra checks these layers on every signup and groups related signups, so a bulk campaign stands out even when each account looks plausible alone.

### Why do people create fake accounts?

Mostly for money or reach. Fake accounts farm signup bonuses and referrals, post spam and scams, write fake reviews, move stolen funds, test stolen cards, multiply free trials and API quotas, and inflate follower or vote counts. Many are also sold to other fraudsters, especially once they have aged or passed verification.

### What is a synthetic identity?

A synthetic identity combines real and invented details, such as a real identification number with a made-up name and date of birth, to create a person who does not exist. It can pass basic onboarding checks and build a credit or trust history over time. It matters most in fintech and lending, where it is used to obtain credit that is never repaid.

### Does phone verification stop fake accounts?

It slows casual abuse but not organized operations. Virtual number services supply codes cheaply, and each verification text costs you money. Attackers can even abuse the SMS step itself for toll fraud. Assessing the signup first and sending codes only to low-risk requests cuts both fake accounts and messaging costs.

### Should I block suspicious signups or let them in and watch?

Block when the evidence is clear, such as automation from an emulator in a signup surge. When it is mixed, let the account exist with limits: no rewards, no messaging or lower limits until it shows normal activity. Starting in observe-only mode shows which approach suits your traffic before you enforce anything.

---
Kavra Lab: bot and fraud detection that explains every decision. Book a demo: https://kavralab.com/contact/