Websites identify visitors through browser fingerprints, which are actually combinations of more than a dozen signals. This article explains 16 common browser fingerprint dimensions, including IP, time zone, UA, Canvas, WebGL, and AudioContext, and how they work.
When you visit a website, it can silently collect many details exposed by your browser: your operating system, screen resolution, installed fonts, the graphics card used for rendering, and more. Individually, these details may seem ordinary, but combined they can form a fairly unique “browser fingerprint.” Even if you clear cookies or use incognito mode, a website may still recognize you through it.
For teams operating multiple accounts, understanding each fingerprint dimension helps explain how environment parameters should be configured and why certain parameters must remain consistent with one another. This article walks through 16 of the most common browser fingerprint types.
Network and location fingerprints
IP address. Every time you connect to the internet, your service provider assigns an IP address to your device. It works somewhat like a mailing address in the physical world, ensuring data reaches the correct destination. Websites use it to determine the visitor’s network origin for geolocation, content customization, and security checks such as detecting unusual logins. IP addresses come in IPv4 format (for example, 192.168.1.1) and IPv6 format (for example, 2001:0db8:85a3::8a2e:0370:7334).
Geolocation. An IP address can be used to roughly infer a visitor’s country, region, city, and sometimes more specific location information. Websites use this data to provide localized content, language preferences, and regional advertising.
Time zone and local time. A time zone can be inferred from the IP address or read directly through browser scripts. The key risk is consistency: if the IP indicates the United States while the browser time zone is set to Beijing time, the mismatch can easily be flagged as abnormal by anti-fraud systems.
Basic browser fingerprints
User Agent (UA). This is a string that the browser sends to a website containing information such as browser type, version, and operating system. For example, Chrome on Windows includes engine and system version information in its UA. Websites use the UA to serve compatible content and interfaces.
Browser language. This is the preferred language configured in the browser and sent to the server through HTTP request headers. One important point is that the browser interface language and the externally exposed language fingerprint can be different. As long as the language fingerprint matches the proxy IP region, having the interface displayed in Chinese does not affect environment consistency.
Fonts. Each device has a different set of installed fonts, and browsers can use scripts to enumerate available fonts. A device with many non-standard fonts may have a font list that becomes a strong identifying signal on its own.
Screen resolution. This includes both the physical screen resolution and the browser viewport size after toolbars and tab bars are excluded. Display combinations differ noticeably across devices, making resolution a common fingerprint dimension.
Hardware concurrency and device memory. Browsers can expose the number of logical CPU cores through hardwareConcurrency and approximate memory through deviceMemory. The latter is usually reported in powers of two and capped at 8 GB as part of the browser’s privacy protection design.
Do Not Track. Users can enable a “Do Not Track” flag in browser settings. Interestingly, because most users leave this setting unchanged, enabling it can actually make you stand out among default users and become an identifying feature itself.
Graphics and media fingerprints
Canvas fingerprint. A website can ask the browser to draw graphics and text on a Canvas and then analyze the rendering result. Differences in screens, fonts, and graphics cards create tiny pixel-level variations across devices. Combined, these variations form a fairly unique “digital fingerprint,” making Canvas one of the stronger fingerprinting methods.
WebGL and WebGPU. These graphics-related APIs allow web pages to use the GPU to render 3D graphics. Websites can read hardware information such as the graphics card model through them, creating fingerprints at the hardware level. You can see their effect when playing 3D games in a browser, but they are also important sources of fingerprint data.
Media device fingerprint. The MediaDevices API can provide information about devices such as cameras and microphones, including device type, name, resolution, and connection method. For privacy reasons, modern browsers usually require user permission before exposing this information.
AudioContext fingerprint. AudioContext is an API used by browsers to process audio. Subtle differences in how different devices’ audio hardware and software process sound can be captured and turned into an “audio fingerprint,” which also has relatively high distinguishing power.
Layout and other fingerprints
ClientRects. When rendering a web page, the browser continuously calculates the position and size of page elements. The results can vary by device, browser, and window size. Websites can combine this layout information into another set of identifying signals.
SpeechVoices. This fingerprint is based on the speech-synthesis voices available on a device, including different languages, accents, and speakers. Voice configurations are unlikely to be exactly identical across devices, so they can also provide identifying value.
Understand fingerprints to configure matching environments
Among the 16 dimensions above, IP, time zone, language, UA, Canvas, WebGL, and AudioContext are some of the most important for multi-account operations. When configuring a browser environment, the goal is not to make every parameter “special,” but to keep the whole set consistent. If the proxy IP is in the United States, the time zone, language, and UA should all point to the same region. Contradictory parameters are often more conspicuous.
When creating an environment, PurpleMark’s web version covers the main dimensions above: operating system and Chromium engine version, User-Agent and resolution, browser language, time zone and geolocation (following the IP or customized), font list, WebGL metadata, WebGPU, WebRTC and proxy UDP, plus controls for Canvas, AudioContext, media devices, ClientRects, SpeechVoices, Do Not Track, hardware acceleration, MAC address, and port-scan protection. A simple configuration principle is to use the target region of the business account as the baseline, keep network, system, language, and advanced fingerprint settings aligned with that region, and keep the environment stable over time.
It is important to note that environment isolation reduces the risk of accounts being linked because they share the same environment. Whether an account remains stable over the long term still depends on content quality and whether operations follow platform rules. Fingerprint settings are only one part of the picture.
Frequently asked questions
Can a browser fingerprint change?
Yes. Changing device settings, installing new software, or updating the browser or operating system can all change the fingerprint. That is why once a multi-account environment is configured, it is generally better to keep it stable and avoid frequent changes.
Can I still be recognized after clearing cookies?
Possibly. Cookies are only one identification method. Fingerprints such as Canvas, WebGL, and AudioContext do not depend on cookies, so they can still exist after cookies are cleared. This is one reason why “incognito mode does not mean invisible.”
Do all websites collect browser fingerprints?
Not every website actively collects them, but major social, e-commerce, and advertising platforms commonly use fingerprinting for risk control, anti-fraud, and ad targeting. The impact is limited for ordinary users but can be significant for multi-account operators.
