Why new TikTok accounts get banned

It was made in a batch

The cheapest accounts on the market are created hundreds at a time: same script, same device farm, same IP range, same timing pattern, phone numbers pulled from the same pool of virtual numbers.

That leaves a fingerprint. Platforms don't need to identify each account individually — they cluster. When one account in a cluster is reported or behaves badly, the whole cluster becomes suspect, and the rest are removed in a sweep. This is why a $20 account can work perfectly for a week and then disappear alongside four hundred others.

It was made on an emulator

An emulator can render the app but it cannot convincingly produce what a real phone produces: accelerometer and gyroscope noise, battery discharge curves, screen touch pressure and timing, camera and microphone hardware identifiers, radio state changes as you move between cells.

Detection here has been mature for years. An account whose entire history shows a device that never once moved, never rotated, and never changed network is not a hard case.

It behaved like software on day one

New accounts get watched closely. A real person opens the app, scrolls for a while, watches some videos to the end, likes a few things, maybe follows two or three accounts, and closes it.

Accounts that instead follow fifty profiles in ten minutes, post immediately after signup, comment the same phrase repeatedly, or send DMs before they have any followers are doing something a human almost never does. Each of those on its own is a small signal; together they're a decision.

The phone number was recycled

Virtual and rented numbers get reused across thousands of signups. A number that has verified two hundred accounts is worth very little as a trust signal, and it links every account that used it.

It also means you can lose the account with no recourse: if you can't receive an SMS on that number any more, and you don't hold the account's email address, recovery is closed to you.

Profile and content trip an automated filter

A brand-new account with a link in the bio, a username that reads like a keyword string, a stock-image avatar and a first post that is an obvious ad is the exact shape of an account created to promote something.

None of these is banned by itself. Together, on a zero-history account, they're the profile of something to be checked.

It was logged into from five places at once

A brand-new account that appears from three countries and four devices in its first day is describing something that does not happen to real people. Neither does an account that only ever connects through a datacentre IP, or one that changes its apparent country every session.

This is the part buyers most often cause themselves. The account is created cleanly in one place, and then it is opened on a laptop browser, a phone, an automation tool and a shared VPN within the same afternoon — and the platform reads the pattern as an account changing hands.

Pick one device and one ordinary connection for the first week, and add anything else slowly. It costs nothing and removes the most common reason a correctly-made account still dies young.

What actually reduces the risk

Start from a clean creation: a real device, a real SIM, an ordinary residential connection, one account at a time.

Give it a few normal days before you ask anything of it. Watch content in the niche you care about. Like a handful of things. Follow a few accounts. This is the whole of what people call "warming" and it is not complicated — it just takes hours you have to actually spend.

Hold the account's email address. If the account is ever locked or challenged, the email is what gets it back.

And expect a failure rate. Even accounts created perfectly get caught in sweeps sometimes. That's why any serious service replaces them instead of arguing about it.