Browser fingerprint and accounts: what links them and what doesn't

StockRush··7 min read
Browser fingerprint and accounts: what links them and what doesn't

A browser fingerprint is the combination of technical traits a website can read from your browser and device (fonts, graphics output, screen, language, time zone, hardware hints) that together are specific enough to recognise the same machine without a cookie. Platforms use it to tie accounts together: if two accounts keep showing up with the same fingerprint, they get treated as one operator. That is how browser fingerprint accounts end up linked, and it is why a login on a purchased account can end badly even when the password is correct.

What a browser fingerprint is made of

No single value in a fingerprint is secret. Every page you open asks the browser for these details to render correctly, and the fingerprint is what you get when you collect them in one place and hash the result. They fall into a few layers.

The browser layer is the obvious one: user-agent string, browser version, language list, time zone, installed plugins and the do-not-track flag. The display layer adds screen size, pixel ratio, colour depth and the fonts installed on the system. Fonts are surprisingly revealing, because the list depends on the operating system and on whatever software the owner has installed.

The rendering layer is where most of the uniqueness comes from. A script can ask the browser to draw a hidden image with canvas or WebGL and read the pixels back. The result depends on the GPU, the driver, the operating system and the font renderer, so two machines rarely draw it identically. Audio processing behaves the same way. On top of that come hardware hints: the number of CPU cores, device memory, touch support.

Strictly speaking, the network is a neighbour rather than a part of the fingerprint. The IP address, the TLS handshake and the order of HTTP headers are visible to the server as well, and platforms read them alongside the browser data.

A platform stores the fingerprint, or a set of its components, with each login. When a new account arrives with values that already belong to another account, the two land in the same cluster. One match rarely triggers anything on its own. What matters is the pile-up: same fingerprint, same IP range, same payment details, same login rhythm. Each shared signal raises the odds that one person runs both accounts, and once a cluster exists, a ban on one account tends to drag its neighbours down too.

Rarity cuts both ways. A default Chrome on a popular laptop produces a fingerprint shared by a huge number of strangers, so a match means little. A heavily customised browser with unusual fonts and odd screen settings is far easier to single out, and a match on it means a lot.

Stability matters as well. A fingerprint that changes on every visit looks like a tool, not a person. A user-agent claiming Windows while the fonts and rendering say macOS is a contradiction a risk system can score directly.

Fingerprint, cookies, IP and session are four different things

Cookies are stored by the site in your browser, and you can delete them. A fingerprint is computed from what the browser reveals, so there is nothing to delete. The IP address belongs to the connection and changes when you switch networks. A session is the token proving you already logged in, which is covered in account cookies and tokens explained.

People mix them up because all four show up in one login event. The platform sees a session or a password, a fingerprint, and an IP from some country, then compares all of it with the history of that account. A correct password combined with a fingerprint and IP never seen for that account is the classic trigger for a verification screen.

What platforms compare, and what changing it does

Platforms do not publish their weights, so treat this table as a map of categories, not a specification of any one system.

SignalWhat it revealsDoes changing it help?
User-agent stringClaimed browser and operating systemNot on its own; a mismatch with other signals is worse than the original
Cookies and local storageEarlier visits from this browserClearing removes history, not the fingerprint
Canvas, WebGL, audioGPU, driver, rendering stackRandom noise often looks inconsistent and can itself stand out
Fonts and screenOperating system and setupHard to change consistently by hand
Time zone and languageThe location story of the userMust agree with the IP location
IP and connectionNetwork originMatters, but is one signal among many
Hardware hintsCores, memory, touchMust fit the device the browser claims to be

The last column is the point. Patching one row rarely helps, because the rows are checked against each other.

What is worth doing and what is useless

Incognito mode is useless here: it throws away cookies when the window closes and leaves canvas, fonts and hardware exactly as they were. Clearing cookies has the same limit. A user-agent switcher extension changes one string and creates a contradiction with everything else. A VPN changes the IP and nothing inside the browser.

What does help is boring consistency.

  1. Give each purchased account its own browser profile and keep using that profile. Do not log into several accounts from one profile.
  2. Match the time zone and language to the place the account normally logs in from.
  3. Do not jump between browsers, devices or settings in the first days of using the account.
  4. If the account asks for verification, complete it from the same environment instead of trying another one.

A dedicated tool can build a coherent profile instead of patching single values, and what an antidetect browser actually fakes shows how far that goes and where it stops.

One limit deserves a plain statement. If a platform forbids several accounts per person, a clean fingerprint does not make that allowed, and a linked-account ban is not something support will casually reverse. Fingerprints are also only one cause of early bans; the broader picture is in why purchased accounts get banned so fast.

What the catalog tells you about fingerprints

StockRush lists 860 active offers from 198 shops across 125 services. Those numbers describe the catalog, not fingerprint risk, but they have a practical consequence: for a popular service you will usually have several sellers to pick from, and accounts from different sellers are not made the same way.

Before buying, ask the seller how the accounts were registered and whether a batch was created or logged in from a single browser profile. Accounts from one batch may already share a fingerprint before you touch them, which links them to each other whatever you do afterwards. The listing will not answer that, only the seller can. You can see how many sellers offer the same service in the product catalog.

Short answers

Does incognito mode change my browser fingerprint?

No. Incognito discards cookies and history when the window closes. Fonts, screen, GPU rendering and hardware hints stay the same, so a site sees the same fingerprint as in a normal window.

Can accounts be linked if they use different IPs?

Yes. The IP is one signal. If the fingerprint, payment details, recovery contacts and behaviour match, a different IP only removes one of the links between the accounts.

Is a rare fingerprint better than a common one?

Usually the opposite. A common setup blends in with many users, while a rare one is easy to recognise, so a repeated match is strong evidence. Consistent and ordinary beats unique.

Find it in the catalog

Related Posts