How Web Browsers Work on Phones and PCs: From Address to Page, Step by Step

A web browser turns an address into a page in a few steps that take a fraction of a second. It looks up the site's IP address through DNS, opens a secure HTTPS connection and checks the site's certificate, then requests the page with HTTP. The server sends back HTML, which the browser's rendering engine parses into a structure called the DOM, styles with CSS, brings to life with JavaScript, lays out and finally paints to the screen. Chrome, Edge, Opera and Brave use the Blink engine and V8 for JavaScript; Firefox uses Gecko; Safari uses WebKit. On phones, the steps are the same, but browsers save data, battery and memory more aggressively, and on iPhone, nearly all browsers use WebKit. Browsers also keep each site in a sandbox, cache files to load faster, and store cookies that keep you signed in.

Ethan's niece asked him a question for a school project: when she typed a website's name on her phone, how did the page arrive so fast, and why did the same site look different on her laptop? Ethan realized he used a browser all day without knowing what it did. He asked Jake, who drew the journey on a napkin: a name becomes a number, a locked connection opens, files travel across the world, and the browser assembles them like a recipe into the page. By the end of lunch, his niece had a diagram for her project, and Ethan understood why updates matter, why some pages are slow and why his phone sometimes reloaded tabs he thought were open. This page explains how browsers work on phones and PCs, step by step.

The journey from address to page

1. You type an address or tap a link, such as www.example.com.
2. DNS lookup: the browser finds the server's IP address for that name.
3. Connection: the browser connects to the server and sets up an encrypted HTTPS connection, checking the site's certificate.
4. Request: the browser asks for the page with an HTTP request.
5. Response: the server sends the HTML document, and the browser requests the images, styles and scripts it refers to.
6. Parsing: the browser reads the HTML into the DOM and the CSS into style rules.
7. Scripting: JavaScript runs, often changing the page or loading more content.
8. Layout and paint: the browser works out where everything goes and draws it on screen.

All of this usually happens in well under a second for a typical page, and many steps overlap: the browser starts showing parts of the page before everything has arrived. Each step is also a place where things can go wrong, which is why understanding them helps explain slow pages, security warnings and errors.

A simple way to picture it

A helpful way to explain browsers to children, or anyone new to the idea, is a restaurant. Typing an address is like naming a restaurant; DNS is the phone book that gives its street address. HTTPS is a sealed envelope that only you and the kitchen can open, with an ID card proving it is really that restaurant. The HTTP request is your order. HTML is the recipe for the dish's structure, CSS is how it is plated and decorated, and JavaScript is the waiter who adds things at the table when you ask. The browser is the chef in your own kitchen, assembling everything into the meal on your screen.

The analogy is not perfect, but it captures the key point: the page you see is built on your device from instructions and ingredients sent by the website, which is why your device, browser and settings affect what you see as much as the site does.

It also explains why two people can see the same site slightly differently: different devices, browsers and settings assemble the same recipe in slightly different kitchens.

The address bar: search or website?

Before any of the steps above, the browser decides what you meant. The address bar, which Chrome calls the omnibox, accepts both addresses and searches. If what you type looks like an address, such as a name with a dot and a known ending, the browser treats it as a website; otherwise, it sends it to your default search engine. Suggestions appear as you type, drawn from your history, bookmarks and the search engine.

That is why a typo can take you to a search results page rather than an error, and why typing an exact address is safer than clicking a search result when you want a specific site such as your bank, since scammers sometimes buy ads that look like real sites in search results. Bookmark the sites you visit for money and accounts.

You can change the default search engine in the browser's settings, under Search or Address bar and search, if the results you get are not the ones you want.

Step 1: DNS turns names into addresses

Computers on the internet find each other by IP address, numbers such as 203.0.113.10 for older IPv4 addresses or longer strings for IPv6. People prefer names. The Domain Name System, DNS, is the internet's directory that turns names into addresses. When you type a name, the browser asks a DNS resolver, usually run by your internet provider or a public service, which finds the answer by asking a chain of servers responsible for each part of the name, then returns the address.

Answers are cached at several levels, in the browser, the operating system and the resolver, for a period set by the site called the time to live, so repeat visits skip most of the lookup. Our guide to what DNS TTL means explains why changes to a site's address can take time to reach everyone. Modern browsers can also encrypt DNS lookups with DNS over HTTPS, so others on the network cannot see which names you look up.

When DNS fails, every site seems unreachable even though your connection is fine, which is a common cause of the frustrating no internet message on a network that otherwise works.

Step 2: a secure HTTPS connection

Once the browser knows the address, it opens a connection to the server, traditionally over TCP, or over the newer QUIC protocol used by HTTP/3. For HTTPS sites, which are nearly all sites today, the browser and server then perform a TLS handshake: they agree on encryption keys, and the server presents a certificate proving it owns the name. The browser checks that the certificate is valid, current and issued by a trusted authority.

If the check passes, everything that follows is encrypted, so others on the network, including public Wi-Fi users, cannot read or change it. If it fails, the browser shows a warning that the connection is not private. That warning is one of the most important safety features a browser has; our guide to public Wi-Fi and HTTPS explains why you should never click past it on an unfamiliar network.

The padlock or similar icon in the address bar shows that the connection is encrypted, not that the site is trustworthy; scam sites can use HTTPS too. Check the site's name carefully as well, especially for banking and shopping.

Step 3: requests and responses

Over the secure connection, the browser sends an HTTP request asking for the page, along with information such as which browser it is, which languages you prefer and any cookies the site set earlier. The server replies with a status code, such as 200 for success, 301 for a redirect or 404 for not found, plus headers and the page content. Status codes explain many errors you see, like 404 pages and 500 server errors.

Modern versions of HTTP, HTTP/2 and HTTP/3, let the browser fetch many files at once over one connection, which matters because a typical page needs dozens or hundreds of files: images, stylesheets, scripts, fonts and data. The browser discovers most of these as it reads the HTML and requests them as soon as possible.

Redirects and link tracking

Clicking a link does not always go straight to its destination. Many links, especially in emails, ads and social media, pass through one or more redirects first, short stops on other servers that record the click before sending the browser on. You may see the address bar change several times before the page loads. Search engines and email services also wrap links to check them for safety.

Redirects are mostly harmless, but they add a little delay and record your activity. Some browsers strip known tracking parameters from links automatically. Hovering over a link shows its first destination in the bottom corner of the browser, a useful check before clicking links in unexpected emails.

Link shorteners work the same way, sending you to the real address through a redirect, which is why it is worth previewing shortened links from strangers before opening them.

Common HTTP status codes

Codes starting with 4 usually mean something about the request, such as a wrong address; codes starting with 5 mean the server has a problem, and waiting or trying later is usually the only fix from your side.

Developer tools show these codes for every file a page loads.

Browser error messages decoded

Browser errors map neatly onto the steps on this page. DNS_PROBE_FINISHED_NXDOMAIN or server IP address could not be found means the DNS step failed: the name is wrong, the site no longer exists, or DNS is not working on your network. ERR_CONNECTION_TIMED_OUT or ERR_CONNECTION_REFUSED means the connection step failed: the server is down, or a firewall is blocking it. Your connection is not private or NET::ERR_CERT errors mean the HTTPS certificate check failed.

Knowing which step failed tells you where to look. DNS errors on every site point to your network; certificate errors on every site often mean the PC's clock is wrong; an error on one site only usually means the problem is with that site. Restarting the router fixes many DNS and connection errors at home.

Trying the same site on mobile data instead of Wi-Fi is a quick test: if it works there, the problem is your home network, not the site.

Step 4: turning code into a page

The browser's rendering engine turns the downloaded code into what you see. It parses the HTML, the page's structure of headings, paragraphs, links and images, into the Document Object Model, or DOM, a tree that represents every element. It parses the CSS, which describes colors, fonts, sizes and positions, into style rules and matches them to the elements. Together these form the render tree.

Next comes layout, working out the exact size and position of every element for your screen size, then paint, filling in pixels, and compositing, assembling layers, often using the graphics processor, into the final image. When something changes, such as text loading or a menu opening, the browser repeats the needed steps. Pages that change layout a lot as they load feel jumpy, which is why well-built sites reserve space for images and ads in advance.

Responsive design: one site, every screen

Most modern sites use responsive design: the same HTML with CSS rules that change the layout depending on screen width. On a wide laptop screen, a news site might show three columns and a sidebar; on a phone, the same page stacks into a single column with a menu button. The browser applies the rules for your screen as it lays out the page, and reapplies them if you rotate a phone or resize a window.

That is why resizing a desktop browser window narrower often makes a site switch to its phone layout. Browsers on phones can also request the desktop version of a site from their menu, which is useful when a mobile layout hides something you need.

Developer tools include a device mode that previews how a page looks on different phone sizes, a quick way to see responsive design in action.

Images, video and fonts

Pages are made of more than text. Browsers decode images in formats such as JPEG, PNG, and the newer WebP and AVIF, which give similar quality at much smaller sizes, so modern sites load faster. Video uses codecs such as H.264, VP9 and AV1, decoded by the graphics chip where possible to save battery. Web fonts let sites use their own typefaces, downloaded alongside the page.

Media is usually the bulk of a page's size. Browsers delay offscreen images until you scroll, and on phones they may choose smaller image versions for smaller screens. Data saver modes and lighter page versions reduce media further on slow or metered connections.

If videos stutter while other sites are fine, updating the graphics driver often helps, since browsers hand video decoding to the graphics hardware.

Step 5: JavaScript makes pages interactive

JavaScript is the programming language of the web, used for menus, forms, maps, video players, chat, games and whole applications like webmail and online documents. Each browser has a JavaScript engine that runs it: V8 in Chrome, Edge and other Chromium browsers, SpiderMonkey in Firefox, and JavaScriptCore in Safari. Engines compile JavaScript into fast machine code as it runs, which is why web apps can feel as quick as installed programs.

JavaScript can change the DOM, fetch more data in the background and respond to clicks and typing. Heavy scripts, including many ad and tracking scripts, are the main reason pages load slowly or drain phone batteries. Blocking trackers, as browsers' tracking protection does, often makes pages noticeably faster.

Some sites barely work without JavaScript, because the whole page is built by scripts in the browser rather than sent as finished HTML. These single-page apps load one page and then update it as you click, which feels fast once loaded but can be slow to start on older phones.

Web standards: why sites work everywhere

Browsers from different companies can show the same site because they follow shared web standards for HTML, CSS, JavaScript and the protocols beneath them. Standards bodies such as the World Wide Web Consortium, the WHATWG, which maintains the living HTML standard, the IETF for network protocols, and Ecma International for JavaScript publish the rules, and browser makers implement them, increasingly working together on shared test suites so features behave the same everywhere.

When a site works in one browser but not another, it often relies on a feature one engine has not implemented yet, or on non-standard behavior. Updating the browser usually helps, and reporting the problem to the site's owner helps the next visitor.

Browser makers publish their plans for new features openly, so developers can prepare and test sites before changes reach everyone.

Browser engines: Blink, Gecko and WebKit

Because most browsers share one of three engines, a site that works in Chrome usually works in Edge and Brave too, and a flaw in an engine affects every browser built on it. That is why a Chrome security update is often followed within days by updates to Edge and others; our guide to updating Chrome and checking your version covers this. Microsoft Edge switched from its own engine to Chromium in 2020, which is one reason so many browsers now share Blink.

Having only a few engines makes the web more consistent for developers, but critics worry it gives the companies behind them great influence over how the web develops, which is one reason Firefox's independent engine is valued.

A short history of browsers

The first widely used graphical browser, Mosaic, appeared in 1993, followed by Netscape Navigator in 1994 and Microsoft's Internet Explorer in 1995, which dominated the early 2000s. Firefox arrived in 2004 from the Mozilla project that grew out of Netscape. Apple released Safari in 2003, built on WebKit. Google launched Chrome in 2008, with a fast JavaScript engine and separate processes for tabs, ideas that reshaped every browser.

Chrome's open-source base, Chromium, became the foundation for many others, and Google forked WebKit to create Blink in 2013. Microsoft replaced Internet Explorer with Edge in 2015, then rebuilt Edge on Chromium in 2020, and Internet Explorer was retired in 2022. Today, three engines, Blink, Gecko and WebKit, power nearly every browser people use.

The browser wars of the late 1990s and the Chrome era both drove rapid innovation, and today's shared engines are a result of those battles.

Internet Explorer's retirement also ended support for many old intranet sites, which Edge now opens in a compatibility mode for businesses that still need them.

How browsers work on phones

On phones, browsers follow the same steps, but they work within tighter limits on memory, battery, processing power and mobile data. Mobile browsers prioritize the visible part of the page, delay loading images further down until you scroll, and may freeze or discard tabs in the background to free memory, which is why a tab sometimes reloads when you return to it. Sites detect smaller screens and use responsive designs that rearrange content for phones, which is why the same site looks different on a laptop.

On iPhone and iPad, Apple has long required browsers to use its WebKit engine, so Chrome, Firefox and Edge on iOS are WebKit underneath with their own features and sync on top. Apple has allowed other engines in the European Union under the Digital Markets Act, but elsewhere WebKit remains the rule. On Android, browsers use their own engines, and many apps show web pages through Android System WebView, a Chromium-based component updated through the Play Store.

In-app browsers

When you tap a link inside an app such as a social network or email app, the page often opens in an in-app browser rather than your main browser. On Android, these use Chrome Custom Tabs or the system WebView; on iPhone, Safari View Controller or WebKit inside the app. In-app browsers are convenient but may not share your main browser's sign-ins, saved passwords or extensions, and some apps can observe what you do in them.

For banking, shopping and anything sensitive, use the menu option to open the link in your main browser. Many apps also have a setting to always open links in the default browser instead.

In-app browsers usually show the site's address at the top; if they hide it, be extra careful about entering passwords there.

Saving data on phones

On limited mobile data plans, browsers offer ways to use less. Chrome's Lite mode was retired in 2022 as mobile data became cheaper, but phones themselves have data saver settings that limit background data, and many browsers can block images or autoplaying video. Blocking ads and trackers also cuts data noticeably, since they often account for a large share of a page's size.

Downloading articles over Wi-Fi to read later, or using a browser's reading list, avoids using mobile data at all for longer reads. Video is by far the largest data user; lowering streaming quality on mobile saves the most.

Checking your phone's data usage screen shows how much each browser uses, which helps decide whether these savings matter for you.

How browsers keep you safe

Browsers run code from strangers every time you open a page, so they are built with layers of protection. Each site runs in a sandbox, a restricted process that cannot read your files or other sites' data. Chrome's site isolation puts different sites in separate processes, so a flaw exploited on one site is harder to use against another. Same-origin rules stop one site's scripts reading another site's data. Safe Browsing and Microsoft Defender SmartScreen warn about known dangerous sites and downloads.

None of this works well without updates, because attackers look for flaws in the sandbox and engines, and browsers fix several every month. Keeping the browser updated is the single most important security step. Extensions, which run with broad access to your browsing, are the other big risk; install few, from trusted developers, and remove ones you no longer use.

How browsers help accessibility

Browsers build an accessibility tree from the DOM, a simplified description of headings, links, buttons and form fields that screen readers such as Narrator, NVDA and VoiceOver use to read pages aloud and navigate them. Well-built sites that use proper headings, labels and descriptions work far better with screen readers than sites built only for appearance.

Browsers also offer zoom with Ctrl + Plus and Minus, reading modes that simplify pages, read-aloud features, and settings for minimum font size and high-contrast display. These help many people, not only those with disabilities, read the web comfortably.

Browser zoom remembers each site's level, so a hard-to-read site stays enlarged on your next visit.

How extensions fit in

Browser extensions add features by running their own code inside the browser, with permissions you grant when installing them. Depending on those permissions, an extension can read and change pages you visit, block content, manage passwords or add buttons. That power is why ad blockers and password managers are possible, and why a malicious or sold-off extension can spy on everything you do.

Chrome and Edge moved to a newer extension system, Manifest V3, which limits some powerful abilities, changing how ad blockers work. When an extension stops working after an update, our guide to Chrome extensions missing or disabled explains why. Keep extensions few and trusted, and review their permissions.

An extension's permissions are listed on its details page in the browser's extensions settings, where you can also limit some to run only on sites you choose.

Cache, cookies and storage

Browsers store data to make browsing faster and more convenient. The cache keeps copies of images, scripts and styles so repeat visits load quickly. Cookies are small pieces of data sites ask the browser to store and send back, used to keep you signed in, remember a shopping cart or preferences, and, for third-party cookies, track you across sites. Sites can also use local storage and other databases to save larger amounts of data, which web apps use to work offline.

Clearing the cache fixes some page display problems and frees space, especially on phones; our guide to phone storage and browser cache covers that side. Clearing cookies signs you out of sites. Browsers increasingly restrict third-party cookies, and tracking protection, as our guide to tracking prevention in Edge explains, blocks many trackers outright.

Each browser keeps its own cache and cookies, so signing in on one browser does not sign you in on another.

Why sites ask about cookies

Many sites show a banner asking you to accept or manage cookies. These come from privacy laws, particularly in Europe, that require consent before sites use cookies for advertising and tracking. Essential cookies, which keep you signed in or remember your cart, do not need consent, but tracking cookies do. Choosing reject or manage, where offered, limits tracking without breaking the site in most cases.

Some browsers and extensions can answer these banners automatically, and some sites honor a browser's global privacy signal setting. Whatever you choose, it is stored in a cookie, which is why clearing cookies brings the banners back.

Consent banners are separate from the browser's own cookie settings, which control what is stored regardless of what a banner says.

What private browsing does and does not do

Private browsing, called InPrivate in Edge and Incognito in Chrome, opens a separate session that does not save history, cookies or form data after you close it. It is useful on shared computers and for keeping a gift search away from the family's history. It does not make you anonymous: websites, your employer or school network, and your internet provider can still see your activity, and signing in to a site identifies you as usual.

Private windows also start without your cookies, which makes them a handy test when a site misbehaves: if it works in a private window, clearing that site's cookies and cache in your normal window usually fixes it.

Browser sync across devices

Signing in to a browser with a Google, Microsoft, Mozilla or Apple account syncs bookmarks, history, passwords, open tabs and settings across your phone and PCs. That is how a tab opened on your phone can be continued on your laptop, and how a new PC can have all your bookmarks within minutes. Sync data is encrypted in transit and, with some browsers, can be encrypted with a passphrase only you know.

Sync is convenient but means your browsing data is stored with the account provider. On shared or public computers, do not sign in to sync; use a private window instead and sign out of sites when finished.

Turning sync off on a device keeps its data local while leaving your account's synced copy intact for other devices.

Why some pages are slow

A slow page is usually slow for one of a few reasons. The connection may be slow or congested, especially on mobile data or busy Wi-Fi. The server may be far away or overloaded. The page may be heavy, with large images, videos and many scripts, especially ads and trackers. Or the device may be short of memory or processing power, struggling with many open tabs. Browsers' developer tools show exactly which files took longest, though for everyday use, trying another network or closing tabs tells you most of what you need.

If every site is slow or crashing in every browser, the problem is likely the PC, its network or security software rather than the sites; our guide to all browsers freezing or crashing walks through the shared causes.

Tabs and memory on PCs

Each open tab can use a noticeable amount of memory, since every site runs in its own process with its own copy of scripts and data. Dozens of open tabs can slow down even a capable PC. Browsers now manage this automatically: Edge's sleeping tabs and Chrome's Memory Saver put inactive tabs to sleep after a while, freeing memory, and wake them when you click them, which is why a tab may briefly reload.

You can adjust these in the browser's performance or system settings, including exceptions for sites that must keep running, such as music or chat. Closing tabs you no longer need, and using bookmarks or a reading list for later, keeps both the browser and the PC responsive.

On phones, the same idea applies even more strictly, since phones have less memory and the system may close the whole browser in the background.

Look inside with developer tools

1. Open a page in Chrome, Edge or Firefox and press F12, or Ctrl + Shift + I.
2. Select the Network tab, then reload the page with F5.
3. Watch the list fill with every file the page requests, with its type, size and time.
4. Select a file to see its request and response headers, including the status code.
5. Open the Elements tab to see the live DOM, and hover over items to highlight them on the page.
6. Close the tools with F12 when finished; nothing you view changes the site.

Developer tools are built for web developers, but they are also a fascinating way to see the steps on this page happening for real: dozens of requests, scripts from many companies, and the page's structure behind the design. The bottom of the Network tab shows how many requests a page made and how long it took.

The Console tab shows error messages from the page's scripts, which explains many broken buttons and forms; the Application tab shows the cookies and storage a site has saved on your device.

Offline pages and web apps

Browsers can show pages without a connection in a few ways. A page already in the cache may load from it. Progressive web apps, sites that can be installed like apps from the browser's menu, use a script called a service worker to keep working offline, which is how some email, notes and music sites work on a plane. And you can save pages yourself for reading later; our guide to saving webpages for offline reading covers each method.

To install a web app, look for an install icon in the address bar or the browser menu's Apps or Install option. Installed web apps open in their own window and appear in Start like other apps.

Offline support depends on the site; most ordinary websites still need a connection, and only sites built as web apps work fully offline.

What sites learn about your browser

Every request tells a site something about your browser: its name and version, your language, your screen size and whether you have visited before through cookies. Scripts can learn much more, such as installed fonts, graphics details and time zone, and combine them into a fingerprint that can identify a browser even without cookies. Our guide to browser fingerprinting explains how it works and what reduces it.

This is part of the trade-off of the modern web: the same features that let sites adapt to your device can be used to track it. Privacy-focused browsers and tracking protection settings limit what scripts can learn.

Reducing what you install, using a mainstream browser with default settings and turning on tracking protection makes a browser blend in with many others, which is the practical defense for most people.

Choosing a browser

With most browsers sharing three engines, the differences are mainly in features, privacy defaults and ecosystem. Chrome integrates with Google accounts and services. Edge integrates with Windows and Microsoft accounts and includes features such as vertical tabs and Immersive Reader. Firefox is the main browser not built on Chromium, with strong privacy options. Safari suits Apple devices with efficient battery use. Brave blocks ads and trackers by default.

Any of them, kept updated, is safe for everyday use. Choose based on which accounts you use and which features you value, and set it as default in Windows so links open in it; our guide to changing your default browser covers the steps.

Switching browsers is easy: most import bookmarks, history and passwords from the browser you used before during setup, so you can try one for a week without losing anything.

Keep one browser as your main one to avoid scattered passwords and bookmarks.

What this page said before

The original version of this page, from 2020, was a beginner's guide to how browsers work on mobile phones. This rewrite covers browsers on phones and PCs together, following the full journey from typing an address to seeing a page, with DNS, HTTPS, HTTP, rendering, JavaScript, engines, mobile differences, security, storage and why pages are sometimes slow.

The basics have stayed the same since browsers began, but the scale has changed enormously: pages that once had a few files now load hundreds, and browsers have become some of the most complex software on any device.

How browsers work: frequently asked questions

Ethan's niece got top marks for her diagram, a napkin sketch redrawn neatly with arrows from a name to a number, a lock, a request and a finished page. Ethan now understands why his browser updates so often, why the padlock matters on cafe Wi-Fi, and why his phone's tabs reload. It made the browser feel less like magic and more like a very fast, very careful assistant.

Revision note. Written October 3, 2026, rewriting our 2020 beginner's guide to how browsers work on phones into a complete explanation for phones and PCs, from DNS and HTTPS to rendering, engines, security and speed. May every page you open arrive quickly and safely.