What happens when you type a web address? A page's journey in 5 steps

Step 1: find the server's address (DNS)

You type mybusiness.com. Your browser has no idea where that is: on the Internet, machines are found by a numeric address (an “IP address”, for example 15.237.40.212). So it asks a directory, the DNS, which address is hiding behind that name. The answer comes back within milliseconds, often from the memory of your home router or your Internet provider, which has already looked it up for someone else.

This is where the domain name plays its one and only role: a line in a directory, saying “this name is at this address”. If that line is missing or wrong, the journey stops right there, with a “site not found” message.

Step 2: establish the connection

Address in hand, the browser knocks on the server's door. The two machines agree on how to talk to each other — this is the “handshake”. This step depends on physical distance: a server in Paris answers a visitor in London in around ten milliseconds; a server in California, in around a hundred. Every round trip counts, and there are several.

Step 3: secure the conversation (HTTPS)

Before exchanging any content at all, the server presents its certificate, the proof that it really is the server for mybusiness.com. The browser checks it, then the two agree on an encryption key that nobody else knows. From then on, everything that travels is unreadable to intermediaries — that's the padlock. HTTPS explained for beginners covers this step in detail.

Step 4: request the page, and receive it

The browser finally sends its request: “give me the home page”. What the server does at that moment determines a good part of the website's speed:

The page arrives as text — the HTML — describing the structure: a heading, paragraphs, an image here, a button there. But it doesn't contain the images themselves, nor the fonts, nor the styling: it only refers to them.

Step 5: assemble and display

The browser reads the HTML and discovers everything it is still missing: the stylesheet, the fonts, every image, every script. It goes back to fetch each of these items — often dozens of them — and starts drawing the page as they arrive. That is why a page sometimes appears first without images, or with a temporary font.

This final step explains most slow websites: not the server, but the weight and number of the things that have to be fetched. A 5 MB photo taken on a phone and dropped onto the page as is weighs more than everything else combined. Why a website is slow picks up on this point.

What Leopar does for you

Every step of this journey is taken care of: the domain name is registered and its A record set (step 1), your website is served from a server in Europe (step 2), HTTPS is enabled and renewed automatically (step 3), the page is a static file served as is, with no database to query (step 4), and photos are delivered at the size actually needed rather than as originals (step 5).

In other words, you don't need to understand this journey to have a fast website — but now that you do, you know why it is. Web hosting explained simply goes a little further into the server itself.

Frequently asked questions

Why is the second visit faster than the first?
The browser keeps in memory (the “cache”) the images, fonts and styles it has already received. On the second visit, it only has the HTML left to request.

Where is my website physically?
On a server, that is, a computer that is permanently switched on in a data centre. Its location matters for speed: for European visitors, a server in Europe is preferable.

What do “error 404” and “error 500” mean?
They are the server's answers at step 4. 404: the requested page doesn't exist. 500: the server crashed while trying to build it — typical of a dynamic website, impossible on a static one.

Does all of this also happen on mobile?
In exactly the same way, with a connection that is often less stable: every round trip costs more. Hence the importance of lightweight pages.